Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- README.md +67 -0
- adapter_config.json +34 -0
- adapter_model.safetensors +3 -0
- added_tokens.json +24 -0
- all_results.json +12 -0
- eval_results.json +7 -0
- merges.txt +0 -0
- special_tokens_map.json +31 -0
- tokenizer.json +3 -0
- tokenizer_config.json +212 -0
- train_results.json +8 -0
- trainer_log.jsonl +304 -0
- trainer_state.json +2166 -0
- training_args.bin +3 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
- vocab.json +0 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: sail/Sailor2-8B-Chat
|
| 3 |
+
library_name: peft
|
| 4 |
+
license: other
|
| 5 |
+
tags:
|
| 6 |
+
- llama-factory
|
| 7 |
+
- lora
|
| 8 |
+
- unsloth
|
| 9 |
+
- generated_from_trainer
|
| 10 |
+
model-index:
|
| 11 |
+
- name: la-sailor2-8b-10k-4bit_r32_alpha_16_lr1e-4_3ep
|
| 12 |
+
results: []
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 16 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 17 |
+
|
| 18 |
+
# la-sailor2-8b-10k-4bit_r32_alpha_16_lr1e-4_3ep
|
| 19 |
+
|
| 20 |
+
This model is a fine-tuned version of [sail/Sailor2-8B-Chat](https://huggingface.co/sail/Sailor2-8B-Chat) on the la_chat_data_v4 dataset.
|
| 21 |
+
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 0.2371
|
| 23 |
+
|
| 24 |
+
## Model description
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Intended uses & limitations
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training and evaluation data
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training procedure
|
| 37 |
+
|
| 38 |
+
### Training hyperparameters
|
| 39 |
+
|
| 40 |
+
The following hyperparameters were used during training:
|
| 41 |
+
- learning_rate: 0.0001
|
| 42 |
+
- train_batch_size: 1
|
| 43 |
+
- eval_batch_size: 1
|
| 44 |
+
- seed: 42
|
| 45 |
+
- gradient_accumulation_steps: 4
|
| 46 |
+
- total_train_batch_size: 4
|
| 47 |
+
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 48 |
+
- lr_scheduler_type: cosine
|
| 49 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 50 |
+
- num_epochs: 2.0
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 55 |
+
|:-------------:|:------:|:-----:|:---------------:|
|
| 56 |
+
| 0.5147 | 0.6649 | 10000 | 0.4314 |
|
| 57 |
+
| 0.3106 | 1.3299 | 20000 | 0.3119 |
|
| 58 |
+
| 0.2238 | 1.9948 | 30000 | 0.2371 |
|
| 59 |
+
|
| 60 |
+
|
| 61 |
+
### Framework versions
|
| 62 |
+
|
| 63 |
+
- PEFT 0.11.1
|
| 64 |
+
- Transformers 4.46.1
|
| 65 |
+
- Pytorch 2.4.1+cu121
|
| 66 |
+
- Datasets 2.20.0
|
| 67 |
+
- Tokenizers 0.20.3
|
adapter_config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "sail/Sailor2-8B-Chat",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layer_replication": null,
|
| 10 |
+
"layers_pattern": null,
|
| 11 |
+
"layers_to_transform": null,
|
| 12 |
+
"loftq_config": {},
|
| 13 |
+
"lora_alpha": 16,
|
| 14 |
+
"lora_dropout": 0.0,
|
| 15 |
+
"megatron_config": null,
|
| 16 |
+
"megatron_core": "megatron.core",
|
| 17 |
+
"modules_to_save": null,
|
| 18 |
+
"peft_type": "LORA",
|
| 19 |
+
"r": 32,
|
| 20 |
+
"rank_pattern": {},
|
| 21 |
+
"revision": null,
|
| 22 |
+
"target_modules": [
|
| 23 |
+
"gate_proj",
|
| 24 |
+
"o_proj",
|
| 25 |
+
"v_proj",
|
| 26 |
+
"q_proj",
|
| 27 |
+
"down_proj",
|
| 28 |
+
"up_proj",
|
| 29 |
+
"k_proj"
|
| 30 |
+
],
|
| 31 |
+
"task_type": "CAUSAL_LM",
|
| 32 |
+
"use_dora": false,
|
| 33 |
+
"use_rslora": false
|
| 34 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:97d1b16ebcfaada727bcec7408838910ac48379f7ad9182d68b91100302cf064
|
| 3 |
+
size 369159096
|
added_tokens.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</tool_call>": 151658,
|
| 3 |
+
"<tool_call>": 151657,
|
| 4 |
+
"<|box_end|>": 151649,
|
| 5 |
+
"<|box_start|>": 151648,
|
| 6 |
+
"<|endoftext|>": 151643,
|
| 7 |
+
"<|file_sep|>": 151664,
|
| 8 |
+
"<|fim_middle|>": 151660,
|
| 9 |
+
"<|fim_pad|>": 151662,
|
| 10 |
+
"<|fim_prefix|>": 151659,
|
| 11 |
+
"<|fim_suffix|>": 151661,
|
| 12 |
+
"<|im_end|>": 151645,
|
| 13 |
+
"<|im_start|>": 151644,
|
| 14 |
+
"<|image_pad|>": 151655,
|
| 15 |
+
"<|object_ref_end|>": 151647,
|
| 16 |
+
"<|object_ref_start|>": 151646,
|
| 17 |
+
"<|quad_end|>": 151651,
|
| 18 |
+
"<|quad_start|>": 151650,
|
| 19 |
+
"<|repo_name|>": 151663,
|
| 20 |
+
"<|video_pad|>": 151656,
|
| 21 |
+
"<|vision_end|>": 151653,
|
| 22 |
+
"<|vision_pad|>": 151654,
|
| 23 |
+
"<|vision_start|>": 151652
|
| 24 |
+
}
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.9999002576676919,
|
| 3 |
+
"eval_loss": 0.23712463676929474,
|
| 4 |
+
"eval_runtime": 4319.1383,
|
| 5 |
+
"eval_samples_per_second": 1.548,
|
| 6 |
+
"eval_steps_per_second": 1.548,
|
| 7 |
+
"total_flos": 1.2039888836550205e+19,
|
| 8 |
+
"train_loss": 0.385877049085006,
|
| 9 |
+
"train_runtime": 246042.9971,
|
| 10 |
+
"train_samples_per_second": 0.489,
|
| 11 |
+
"train_steps_per_second": 0.122
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.9999002576676919,
|
| 3 |
+
"eval_loss": 0.23712463676929474,
|
| 4 |
+
"eval_runtime": 4319.1383,
|
| 5 |
+
"eval_samples_per_second": 1.548,
|
| 6 |
+
"eval_steps_per_second": 1.548
|
| 7 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|im_start|>",
|
| 4 |
+
"<|im_end|>",
|
| 5 |
+
"<|object_ref_start|>",
|
| 6 |
+
"<|object_ref_end|>",
|
| 7 |
+
"<|box_start|>",
|
| 8 |
+
"<|box_end|>",
|
| 9 |
+
"<|quad_start|>",
|
| 10 |
+
"<|quad_end|>",
|
| 11 |
+
"<|vision_start|>",
|
| 12 |
+
"<|vision_end|>",
|
| 13 |
+
"<|vision_pad|>",
|
| 14 |
+
"<|image_pad|>",
|
| 15 |
+
"<|video_pad|>"
|
| 16 |
+
],
|
| 17 |
+
"eos_token": {
|
| 18 |
+
"content": "<|im_end|>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
},
|
| 24 |
+
"pad_token": {
|
| 25 |
+
"content": "<|endoftext|>",
|
| 26 |
+
"lstrip": false,
|
| 27 |
+
"normalized": false,
|
| 28 |
+
"rstrip": false,
|
| 29 |
+
"single_word": false
|
| 30 |
+
}
|
| 31 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:82bf42f9364aa867b8c9b839e53c829979092b99b8fab5e8dc932d83f965e815
|
| 3 |
+
size 11421995
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,212 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"151643": {
|
| 6 |
+
"content": "<|endoftext|>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"151644": {
|
| 14 |
+
"content": "<|im_start|>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"151645": {
|
| 22 |
+
"content": "<|im_end|>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
},
|
| 29 |
+
"151646": {
|
| 30 |
+
"content": "<|object_ref_start|>",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false,
|
| 35 |
+
"special": true
|
| 36 |
+
},
|
| 37 |
+
"151647": {
|
| 38 |
+
"content": "<|object_ref_end|>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false,
|
| 43 |
+
"special": true
|
| 44 |
+
},
|
| 45 |
+
"151648": {
|
| 46 |
+
"content": "<|box_start|>",
|
| 47 |
+
"lstrip": false,
|
| 48 |
+
"normalized": false,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false,
|
| 51 |
+
"special": true
|
| 52 |
+
},
|
| 53 |
+
"151649": {
|
| 54 |
+
"content": "<|box_end|>",
|
| 55 |
+
"lstrip": false,
|
| 56 |
+
"normalized": false,
|
| 57 |
+
"rstrip": false,
|
| 58 |
+
"single_word": false,
|
| 59 |
+
"special": true
|
| 60 |
+
},
|
| 61 |
+
"151650": {
|
| 62 |
+
"content": "<|quad_start|>",
|
| 63 |
+
"lstrip": false,
|
| 64 |
+
"normalized": false,
|
| 65 |
+
"rstrip": false,
|
| 66 |
+
"single_word": false,
|
| 67 |
+
"special": true
|
| 68 |
+
},
|
| 69 |
+
"151651": {
|
| 70 |
+
"content": "<|quad_end|>",
|
| 71 |
+
"lstrip": false,
|
| 72 |
+
"normalized": false,
|
| 73 |
+
"rstrip": false,
|
| 74 |
+
"single_word": false,
|
| 75 |
+
"special": true
|
| 76 |
+
},
|
| 77 |
+
"151652": {
|
| 78 |
+
"content": "<|vision_start|>",
|
| 79 |
+
"lstrip": false,
|
| 80 |
+
"normalized": false,
|
| 81 |
+
"rstrip": false,
|
| 82 |
+
"single_word": false,
|
| 83 |
+
"special": true
|
| 84 |
+
},
|
| 85 |
+
"151653": {
|
| 86 |
+
"content": "<|vision_end|>",
|
| 87 |
+
"lstrip": false,
|
| 88 |
+
"normalized": false,
|
| 89 |
+
"rstrip": false,
|
| 90 |
+
"single_word": false,
|
| 91 |
+
"special": true
|
| 92 |
+
},
|
| 93 |
+
"151654": {
|
| 94 |
+
"content": "<|vision_pad|>",
|
| 95 |
+
"lstrip": false,
|
| 96 |
+
"normalized": false,
|
| 97 |
+
"rstrip": false,
|
| 98 |
+
"single_word": false,
|
| 99 |
+
"special": true
|
| 100 |
+
},
|
| 101 |
+
"151655": {
|
| 102 |
+
"content": "<|image_pad|>",
|
| 103 |
+
"lstrip": false,
|
| 104 |
+
"normalized": false,
|
| 105 |
+
"rstrip": false,
|
| 106 |
+
"single_word": false,
|
| 107 |
+
"special": true
|
| 108 |
+
},
|
| 109 |
+
"151656": {
|
| 110 |
+
"content": "<|video_pad|>",
|
| 111 |
+
"lstrip": false,
|
| 112 |
+
"normalized": false,
|
| 113 |
+
"rstrip": false,
|
| 114 |
+
"single_word": false,
|
| 115 |
+
"special": true
|
| 116 |
+
},
|
| 117 |
+
"151657": {
|
| 118 |
+
"content": "<tool_call>",
|
| 119 |
+
"lstrip": false,
|
| 120 |
+
"normalized": false,
|
| 121 |
+
"rstrip": false,
|
| 122 |
+
"single_word": false,
|
| 123 |
+
"special": false
|
| 124 |
+
},
|
| 125 |
+
"151658": {
|
| 126 |
+
"content": "</tool_call>",
|
| 127 |
+
"lstrip": false,
|
| 128 |
+
"normalized": false,
|
| 129 |
+
"rstrip": false,
|
| 130 |
+
"single_word": false,
|
| 131 |
+
"special": false
|
| 132 |
+
},
|
| 133 |
+
"151659": {
|
| 134 |
+
"content": "<|fim_prefix|>",
|
| 135 |
+
"lstrip": false,
|
| 136 |
+
"normalized": false,
|
| 137 |
+
"rstrip": false,
|
| 138 |
+
"single_word": false,
|
| 139 |
+
"special": false
|
| 140 |
+
},
|
| 141 |
+
"151660": {
|
| 142 |
+
"content": "<|fim_middle|>",
|
| 143 |
+
"lstrip": false,
|
| 144 |
+
"normalized": false,
|
| 145 |
+
"rstrip": false,
|
| 146 |
+
"single_word": false,
|
| 147 |
+
"special": false
|
| 148 |
+
},
|
| 149 |
+
"151661": {
|
| 150 |
+
"content": "<|fim_suffix|>",
|
| 151 |
+
"lstrip": false,
|
| 152 |
+
"normalized": false,
|
| 153 |
+
"rstrip": false,
|
| 154 |
+
"single_word": false,
|
| 155 |
+
"special": false
|
| 156 |
+
},
|
| 157 |
+
"151662": {
|
| 158 |
+
"content": "<|fim_pad|>",
|
| 159 |
+
"lstrip": false,
|
| 160 |
+
"normalized": false,
|
| 161 |
+
"rstrip": false,
|
| 162 |
+
"single_word": false,
|
| 163 |
+
"special": false
|
| 164 |
+
},
|
| 165 |
+
"151663": {
|
| 166 |
+
"content": "<|repo_name|>",
|
| 167 |
+
"lstrip": false,
|
| 168 |
+
"normalized": false,
|
| 169 |
+
"rstrip": false,
|
| 170 |
+
"single_word": false,
|
| 171 |
+
"special": false
|
| 172 |
+
},
|
| 173 |
+
"151664": {
|
| 174 |
+
"content": "<|file_sep|>",
|
| 175 |
+
"lstrip": false,
|
| 176 |
+
"normalized": false,
|
| 177 |
+
"rstrip": false,
|
| 178 |
+
"single_word": false,
|
| 179 |
+
"special": false
|
| 180 |
+
}
|
| 181 |
+
},
|
| 182 |
+
"additional_special_tokens": [
|
| 183 |
+
"<|im_start|>",
|
| 184 |
+
"<|im_end|>",
|
| 185 |
+
"<|object_ref_start|>",
|
| 186 |
+
"<|object_ref_end|>",
|
| 187 |
+
"<|box_start|>",
|
| 188 |
+
"<|box_end|>",
|
| 189 |
+
"<|quad_start|>",
|
| 190 |
+
"<|quad_end|>",
|
| 191 |
+
"<|vision_start|>",
|
| 192 |
+
"<|vision_end|>",
|
| 193 |
+
"<|vision_pad|>",
|
| 194 |
+
"<|image_pad|>",
|
| 195 |
+
"<|video_pad|>"
|
| 196 |
+
],
|
| 197 |
+
"bos_token": null,
|
| 198 |
+
"chat_template": "{% set system_message = 'You are a helpful assistant.' %}{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ '<|im_start|>system\n' + system_message + '<|im_end|>\n' }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ '<|im_start|>user\n' + content + '<|im_end|>\n<|im_start|>assistant\n' }}{% elif message['role'] == 'assistant' %}{{ content + '<|im_end|>' + '\n' }}{% endif %}{% endfor %}",
|
| 199 |
+
"clean_up_tokenization_spaces": false,
|
| 200 |
+
"eos_token": "<|im_end|>",
|
| 201 |
+
"errors": "replace",
|
| 202 |
+
"max_length": 3848,
|
| 203 |
+
"model_max_length": 131072,
|
| 204 |
+
"pad_token": "<|endoftext|>",
|
| 205 |
+
"padding_side": "right",
|
| 206 |
+
"split_special_tokens": false,
|
| 207 |
+
"stride": 0,
|
| 208 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 209 |
+
"truncation_side": "right",
|
| 210 |
+
"truncation_strategy": "longest_first",
|
| 211 |
+
"unk_token": null
|
| 212 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.9999002576676919,
|
| 3 |
+
"total_flos": 1.2039888836550205e+19,
|
| 4 |
+
"train_loss": 0.385877049085006,
|
| 5 |
+
"train_runtime": 246042.9971,
|
| 6 |
+
"train_samples_per_second": 0.489,
|
| 7 |
+
"train_steps_per_second": 0.122
|
| 8 |
+
}
|
trainer_log.jsonl
ADDED
|
@@ -0,0 +1,304 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"current_steps": 100, "total_steps": 30076, "loss": 0.818, "learning_rate": 3.324468085106383e-06, "epoch": 0.006649488820546921, "percentage": 0.33, "elapsed_time": "0:12:33", "remaining_time": "2 days, 14:46:44", "throughput": "0.00", "total_tokens": 0}
|
| 2 |
+
{"current_steps": 200, "total_steps": 30076, "loss": 0.5412, "learning_rate": 6.648936170212766e-06, "epoch": 0.013298977641093841, "percentage": 0.66, "elapsed_time": "0:25:10", "remaining_time": "2 days, 14:40:45", "throughput": "0.00", "total_tokens": 0}
|
| 3 |
+
{"current_steps": 300, "total_steps": 30076, "loss": 0.5292, "learning_rate": 9.973404255319148e-06, "epoch": 0.019948466461640762, "percentage": 1.0, "elapsed_time": "0:37:34", "remaining_time": "2 days, 14:09:53", "throughput": "0.00", "total_tokens": 0}
|
| 4 |
+
{"current_steps": 400, "total_steps": 30076, "loss": 0.5203, "learning_rate": 1.3297872340425532e-05, "epoch": 0.026597955282187682, "percentage": 1.33, "elapsed_time": "0:50:51", "remaining_time": "2 days, 14:53:09", "throughput": "0.00", "total_tokens": 0}
|
| 5 |
+
{"current_steps": 500, "total_steps": 30076, "loss": 0.5078, "learning_rate": 1.6622340425531915e-05, "epoch": 0.0332474441027346, "percentage": 1.66, "elapsed_time": "1:04:03", "remaining_time": "2 days, 15:09:35", "throughput": "0.00", "total_tokens": 0}
|
| 6 |
+
{"current_steps": 600, "total_steps": 30076, "loss": 0.4875, "learning_rate": 1.9946808510638297e-05, "epoch": 0.039896932923281524, "percentage": 1.99, "elapsed_time": "1:16:39", "remaining_time": "2 days, 14:46:06", "throughput": "0.00", "total_tokens": 0}
|
| 7 |
+
{"current_steps": 700, "total_steps": 30076, "loss": 0.4978, "learning_rate": 2.3271276595744682e-05, "epoch": 0.04654642174382844, "percentage": 2.33, "elapsed_time": "1:29:39", "remaining_time": "2 days, 14:42:16", "throughput": "0.00", "total_tokens": 0}
|
| 8 |
+
{"current_steps": 800, "total_steps": 30076, "loss": 0.4807, "learning_rate": 2.6595744680851064e-05, "epoch": 0.053195910564375365, "percentage": 2.66, "elapsed_time": "1:42:58", "remaining_time": "2 days, 14:48:32", "throughput": "0.00", "total_tokens": 0}
|
| 9 |
+
{"current_steps": 900, "total_steps": 30076, "loss": 0.5007, "learning_rate": 2.9920212765957452e-05, "epoch": 0.05984539938492228, "percentage": 2.99, "elapsed_time": "1:55:58", "remaining_time": "2 days, 14:39:24", "throughput": "0.00", "total_tokens": 0}
|
| 10 |
+
{"current_steps": 1000, "total_steps": 30076, "loss": 0.4754, "learning_rate": 3.324468085106383e-05, "epoch": 0.0664948882054692, "percentage": 3.32, "elapsed_time": "2:08:51", "remaining_time": "2 days, 14:26:52", "throughput": "0.00", "total_tokens": 0}
|
| 11 |
+
{"current_steps": 1100, "total_steps": 30076, "loss": 0.4943, "learning_rate": 3.6569148936170215e-05, "epoch": 0.07314437702601613, "percentage": 3.66, "elapsed_time": "2:22:10", "remaining_time": "2 days, 14:25:15", "throughput": "0.00", "total_tokens": 0}
|
| 12 |
+
{"current_steps": 1200, "total_steps": 30076, "loss": 0.4928, "learning_rate": 3.9893617021276594e-05, "epoch": 0.07979386584656305, "percentage": 3.99, "elapsed_time": "2:34:59", "remaining_time": "2 days, 14:09:41", "throughput": "0.00", "total_tokens": 0}
|
| 13 |
+
{"current_steps": 1300, "total_steps": 30076, "loss": 0.4829, "learning_rate": 4.3218085106382986e-05, "epoch": 0.08644335466710996, "percentage": 4.32, "elapsed_time": "2:48:31", "remaining_time": "2 days, 14:10:12", "throughput": "0.00", "total_tokens": 0}
|
| 14 |
+
{"current_steps": 1400, "total_steps": 30076, "loss": 0.4797, "learning_rate": 4.6542553191489364e-05, "epoch": 0.09309284348765688, "percentage": 4.65, "elapsed_time": "3:00:46", "remaining_time": "2 days, 13:42:44", "throughput": "0.00", "total_tokens": 0}
|
| 15 |
+
{"current_steps": 1500, "total_steps": 30076, "loss": 0.5052, "learning_rate": 4.986702127659575e-05, "epoch": 0.09974233230820381, "percentage": 4.99, "elapsed_time": "3:14:15", "remaining_time": "2 days, 13:40:53", "throughput": "0.00", "total_tokens": 0}
|
| 16 |
+
{"current_steps": 1600, "total_steps": 30076, "loss": 0.4693, "learning_rate": 5.319148936170213e-05, "epoch": 0.10639182112875073, "percentage": 5.32, "elapsed_time": "3:26:59", "remaining_time": "2 days, 13:23:59", "throughput": "0.00", "total_tokens": 0}
|
| 17 |
+
{"current_steps": 1700, "total_steps": 30076, "loss": 0.4767, "learning_rate": 5.651595744680851e-05, "epoch": 0.11304130994929765, "percentage": 5.65, "elapsed_time": "3:39:45", "remaining_time": "2 days, 13:08:13", "throughput": "0.00", "total_tokens": 0}
|
| 18 |
+
{"current_steps": 1800, "total_steps": 30076, "loss": 0.4823, "learning_rate": 5.9840425531914904e-05, "epoch": 0.11969079876984456, "percentage": 5.98, "elapsed_time": "3:52:05", "remaining_time": "2 days, 12:46:00", "throughput": "0.00", "total_tokens": 0}
|
| 19 |
+
{"current_steps": 1900, "total_steps": 30076, "loss": 0.4631, "learning_rate": 6.316489361702128e-05, "epoch": 0.12634028759039148, "percentage": 6.32, "elapsed_time": "4:05:09", "remaining_time": "2 days, 12:35:26", "throughput": "0.00", "total_tokens": 0}
|
| 20 |
+
{"current_steps": 2000, "total_steps": 30076, "loss": 0.5277, "learning_rate": 6.648936170212766e-05, "epoch": 0.1329897764109384, "percentage": 6.65, "elapsed_time": "4:17:21", "remaining_time": "2 days, 12:12:41", "throughput": "0.00", "total_tokens": 0}
|
| 21 |
+
{"current_steps": 2100, "total_steps": 30076, "loss": 0.4813, "learning_rate": 6.981382978723405e-05, "epoch": 0.13963926523148534, "percentage": 6.98, "elapsed_time": "4:30:11", "remaining_time": "2 days, 11:59:23", "throughput": "0.00", "total_tokens": 0}
|
| 22 |
+
{"current_steps": 2200, "total_steps": 30076, "loss": 0.5026, "learning_rate": 7.313829787234043e-05, "epoch": 0.14628875405203226, "percentage": 7.31, "elapsed_time": "4:42:27", "remaining_time": "2 days, 11:39:05", "throughput": "0.00", "total_tokens": 0}
|
| 23 |
+
{"current_steps": 2300, "total_steps": 30076, "loss": 0.5226, "learning_rate": 7.646276595744682e-05, "epoch": 0.15293824287257918, "percentage": 7.65, "elapsed_time": "4:54:52", "remaining_time": "2 days, 11:21:02", "throughput": "0.00", "total_tokens": 0}
|
| 24 |
+
{"current_steps": 2400, "total_steps": 30076, "loss": 0.5285, "learning_rate": 7.978723404255319e-05, "epoch": 0.1595877316931261, "percentage": 7.98, "elapsed_time": "5:08:03", "remaining_time": "2 days, 11:12:22", "throughput": "0.00", "total_tokens": 0}
|
| 25 |
+
{"current_steps": 2500, "total_steps": 30076, "loss": 0.5183, "learning_rate": 8.311170212765957e-05, "epoch": 0.166237220513673, "percentage": 8.31, "elapsed_time": "5:21:20", "remaining_time": "2 days, 11:04:33", "throughput": "0.00", "total_tokens": 0}
|
| 26 |
+
{"current_steps": 2600, "total_steps": 30076, "loss": 0.5129, "learning_rate": 8.643617021276597e-05, "epoch": 0.17288670933421993, "percentage": 8.64, "elapsed_time": "5:34:07", "remaining_time": "2 days, 10:50:50", "throughput": "0.00", "total_tokens": 0}
|
| 27 |
+
{"current_steps": 2700, "total_steps": 30076, "loss": 0.5081, "learning_rate": 8.976063829787234e-05, "epoch": 0.17953619815476685, "percentage": 8.98, "elapsed_time": "5:46:14", "remaining_time": "2 days, 10:30:35", "throughput": "0.00", "total_tokens": 0}
|
| 28 |
+
{"current_steps": 2800, "total_steps": 30076, "loss": 0.5042, "learning_rate": 9.308510638297873e-05, "epoch": 0.18618568697531376, "percentage": 9.31, "elapsed_time": "5:59:07", "remaining_time": "2 days, 10:18:19", "throughput": "0.00", "total_tokens": 0}
|
| 29 |
+
{"current_steps": 2900, "total_steps": 30076, "loss": 0.5357, "learning_rate": 9.640957446808511e-05, "epoch": 0.19283517579586068, "percentage": 9.64, "elapsed_time": "6:12:05", "remaining_time": "2 days, 10:06:56", "throughput": "0.00", "total_tokens": 0}
|
| 30 |
+
{"current_steps": 3000, "total_steps": 30076, "loss": 0.5407, "learning_rate": 9.97340425531915e-05, "epoch": 0.19948466461640763, "percentage": 9.97, "elapsed_time": "6:24:52", "remaining_time": "2 days, 9:53:40", "throughput": "0.00", "total_tokens": 0}
|
| 31 |
+
{"current_steps": 3100, "total_steps": 30076, "loss": 0.5244, "learning_rate": 9.999714964504067e-05, "epoch": 0.20613415343695454, "percentage": 10.31, "elapsed_time": "6:37:28", "remaining_time": "2 days, 9:38:43", "throughput": "0.00", "total_tokens": 0}
|
| 32 |
+
{"current_steps": 3200, "total_steps": 30076, "loss": 0.5562, "learning_rate": 9.998758599534463e-05, "epoch": 0.21278364225750146, "percentage": 10.64, "elapsed_time": "6:50:07", "remaining_time": "2 days, 9:24:34", "throughput": "0.00", "total_tokens": 0}
|
| 33 |
+
{"current_steps": 3300, "total_steps": 30076, "loss": 0.5273, "learning_rate": 9.997128878619808e-05, "epoch": 0.21943313107804838, "percentage": 10.97, "elapsed_time": "7:02:41", "remaining_time": "2 days, 9:09:41", "throughput": "0.00", "total_tokens": 0}
|
| 34 |
+
{"current_steps": 3400, "total_steps": 30076, "loss": 0.5412, "learning_rate": 9.994826021291056e-05, "epoch": 0.2260826198985953, "percentage": 11.3, "elapsed_time": "7:15:44", "remaining_time": "2 days, 8:58:46", "throughput": "0.00", "total_tokens": 0}
|
| 35 |
+
{"current_steps": 3500, "total_steps": 30076, "loss": 0.5405, "learning_rate": 9.991850337753762e-05, "epoch": 0.2327321087191422, "percentage": 11.64, "elapsed_time": "7:28:01", "remaining_time": "2 days, 8:41:57", "throughput": "0.00", "total_tokens": 0}
|
| 36 |
+
{"current_steps": 3600, "total_steps": 30076, "loss": 0.5049, "learning_rate": 9.988202228846291e-05, "epoch": 0.23938159753968913, "percentage": 11.97, "elapsed_time": "7:40:49", "remaining_time": "2 days, 8:29:05", "throughput": "0.00", "total_tokens": 0}
|
| 37 |
+
{"current_steps": 3700, "total_steps": 30076, "loss": 0.5785, "learning_rate": 9.983882185985808e-05, "epoch": 0.24603108636023605, "percentage": 12.3, "elapsed_time": "7:53:18", "remaining_time": "2 days, 8:14:03", "throughput": "0.00", "total_tokens": 0}
|
| 38 |
+
{"current_steps": 3800, "total_steps": 30076, "loss": 0.5487, "learning_rate": 9.978890791102109e-05, "epoch": 0.25268057518078296, "percentage": 12.63, "elapsed_time": "8:05:18", "remaining_time": "2 days, 7:55:44", "throughput": "0.00", "total_tokens": 0}
|
| 39 |
+
{"current_steps": 3900, "total_steps": 30076, "loss": 0.533, "learning_rate": 9.973228716559209e-05, "epoch": 0.2593300640013299, "percentage": 12.97, "elapsed_time": "8:19:08", "remaining_time": "2 days, 7:50:08", "throughput": "0.00", "total_tokens": 0}
|
| 40 |
+
{"current_steps": 4000, "total_steps": 30076, "loss": 0.5166, "learning_rate": 9.966896725064786e-05, "epoch": 0.2659795528218768, "percentage": 13.3, "elapsed_time": "8:31:18", "remaining_time": "2 days, 7:33:10", "throughput": "0.00", "total_tokens": 0}
|
| 41 |
+
{"current_steps": 4100, "total_steps": 30076, "loss": 0.5366, "learning_rate": 9.959895669567435e-05, "epoch": 0.27262904164242374, "percentage": 13.63, "elapsed_time": "8:44:45", "remaining_time": "2 days, 7:24:38", "throughput": "0.00", "total_tokens": 0}
|
| 42 |
+
{"current_steps": 4200, "total_steps": 30076, "loss": 0.5339, "learning_rate": 9.952226493141765e-05, "epoch": 0.2792785304629707, "percentage": 13.96, "elapsed_time": "8:57:47", "remaining_time": "2 days, 7:13:19", "throughput": "0.00", "total_tokens": 0}
|
| 43 |
+
{"current_steps": 4300, "total_steps": 30076, "loss": 0.4976, "learning_rate": 9.943890228861383e-05, "epoch": 0.2859280192835176, "percentage": 14.3, "elapsed_time": "9:10:53", "remaining_time": "2 days, 7:02:17", "throughput": "0.00", "total_tokens": 0}
|
| 44 |
+
{"current_steps": 4400, "total_steps": 30076, "loss": 0.5331, "learning_rate": 9.934887999659707e-05, "epoch": 0.2925775081040645, "percentage": 14.63, "elapsed_time": "9:24:01", "remaining_time": "2 days, 6:51:19", "throughput": "0.00", "total_tokens": 0}
|
| 45 |
+
{"current_steps": 4500, "total_steps": 30076, "loss": 0.5212, "learning_rate": 9.925221018178728e-05, "epoch": 0.2992269969246114, "percentage": 14.96, "elapsed_time": "9:36:54", "remaining_time": "2 days, 6:38:55", "throughput": "0.00", "total_tokens": 0}
|
| 46 |
+
{"current_steps": 4600, "total_steps": 30076, "loss": 0.4991, "learning_rate": 9.914890586605638e-05, "epoch": 0.30587648574515836, "percentage": 15.29, "elapsed_time": "9:49:23", "remaining_time": "2 days, 6:24:09", "throughput": "0.00", "total_tokens": 0}
|
| 47 |
+
{"current_steps": 4700, "total_steps": 30076, "loss": 0.569, "learning_rate": 9.903898096497441e-05, "epoch": 0.31252597456570524, "percentage": 15.63, "elapsed_time": "10:02:13", "remaining_time": "2 days, 6:11:31", "throughput": "0.00", "total_tokens": 0}
|
| 48 |
+
{"current_steps": 4800, "total_steps": 30076, "loss": 0.5436, "learning_rate": 9.892245028593483e-05, "epoch": 0.3191754633862522, "percentage": 15.96, "elapsed_time": "10:15:13", "remaining_time": "2 days, 5:59:39", "throughput": "0.00", "total_tokens": 0}
|
| 49 |
+
{"current_steps": 4900, "total_steps": 30076, "loss": 0.5028, "learning_rate": 9.879932952616009e-05, "epoch": 0.3258249522067991, "percentage": 16.29, "elapsed_time": "10:27:55", "remaining_time": "2 days, 5:46:14", "throughput": "0.00", "total_tokens": 0}
|
| 50 |
+
{"current_steps": 5000, "total_steps": 30076, "loss": 0.7608, "learning_rate": 9.8669635270587e-05, "epoch": 0.332474441027346, "percentage": 16.62, "elapsed_time": "10:41:07", "remaining_time": "2 days, 5:35:23", "throughput": "0.00", "total_tokens": 0}
|
| 51 |
+
{"current_steps": 5100, "total_steps": 30076, "loss": 0.523, "learning_rate": 9.853338498963272e-05, "epoch": 0.33912392984789297, "percentage": 16.96, "elapsed_time": "10:54:14", "remaining_time": "2 days, 5:24:00", "throughput": "0.00", "total_tokens": 0}
|
| 52 |
+
{"current_steps": 5200, "total_steps": 30076, "loss": 0.5426, "learning_rate": 9.839059703684139e-05, "epoch": 0.34577341866843986, "percentage": 17.29, "elapsed_time": "11:07:16", "remaining_time": "2 days, 5:12:07", "throughput": "0.00", "total_tokens": 0}
|
| 53 |
+
{"current_steps": 5300, "total_steps": 30076, "loss": 0.5387, "learning_rate": 9.82412906464119e-05, "epoch": 0.3524229074889868, "percentage": 17.62, "elapsed_time": "11:20:13", "remaining_time": "2 days, 4:59:51", "throughput": "0.00", "total_tokens": 0}
|
| 54 |
+
{"current_steps": 5400, "total_steps": 30076, "loss": 0.5259, "learning_rate": 9.808548593060681e-05, "epoch": 0.3590723963095337, "percentage": 17.95, "elapsed_time": "11:33:09", "remaining_time": "2 days, 4:47:27", "throughput": "0.00", "total_tokens": 0}
|
| 55 |
+
{"current_steps": 5500, "total_steps": 30076, "loss": 0.5363, "learning_rate": 9.792320387704328e-05, "epoch": 0.36572188513008064, "percentage": 18.29, "elapsed_time": "11:45:30", "remaining_time": "2 days, 4:32:29", "throughput": "0.00", "total_tokens": 0}
|
| 56 |
+
{"current_steps": 5600, "total_steps": 30076, "loss": 0.542, "learning_rate": 9.775446634586584e-05, "epoch": 0.3723713739506275, "percentage": 18.62, "elapsed_time": "11:58:45", "remaining_time": "2 days, 4:21:29", "throughput": "0.00", "total_tokens": 0}
|
| 57 |
+
{"current_steps": 5700, "total_steps": 30076, "loss": 0.4947, "learning_rate": 9.757929606680181e-05, "epoch": 0.37902086277117447, "percentage": 18.95, "elapsed_time": "12:11:04", "remaining_time": "2 days, 4:06:25", "throughput": "0.00", "total_tokens": 0}
|
| 58 |
+
{"current_steps": 5800, "total_steps": 30076, "loss": 0.5584, "learning_rate": 9.73977166360994e-05, "epoch": 0.38567035159172136, "percentage": 19.28, "elapsed_time": "12:23:47", "remaining_time": "2 days, 3:53:09", "throughput": "0.00", "total_tokens": 0}
|
| 59 |
+
{"current_steps": 5900, "total_steps": 30076, "loss": 0.5318, "learning_rate": 9.720975251334929e-05, "epoch": 0.3923198404122683, "percentage": 19.62, "elapsed_time": "12:36:11", "remaining_time": "2 days, 3:38:34", "throughput": "0.00", "total_tokens": 0}
|
| 60 |
+
{"current_steps": 6000, "total_steps": 30076, "loss": 0.5352, "learning_rate": 9.701542901818974e-05, "epoch": 0.39896932923281525, "percentage": 19.95, "elapsed_time": "12:49:24", "remaining_time": "2 days, 3:27:23", "throughput": "0.00", "total_tokens": 0}
|
| 61 |
+
{"current_steps": 6100, "total_steps": 30076, "loss": 0.5229, "learning_rate": 9.681477232689596e-05, "epoch": 0.40561881805336214, "percentage": 20.28, "elapsed_time": "13:02:27", "remaining_time": "2 days, 3:15:27", "throughput": "0.00", "total_tokens": 0}
|
| 62 |
+
{"current_steps": 6200, "total_steps": 30076, "loss": 0.5419, "learning_rate": 9.660780946885397e-05, "epoch": 0.4122683068739091, "percentage": 20.61, "elapsed_time": "13:15:39", "remaining_time": "2 days, 3:04:04", "throughput": "0.00", "total_tokens": 0}
|
| 63 |
+
{"current_steps": 6300, "total_steps": 30076, "loss": 0.5428, "learning_rate": 9.639456832291974e-05, "epoch": 0.418917795694456, "percentage": 20.95, "elapsed_time": "13:28:27", "remaining_time": "2 days, 2:51:05", "throughput": "0.00", "total_tokens": 0}
|
| 64 |
+
{"current_steps": 6400, "total_steps": 30076, "loss": 0.5256, "learning_rate": 9.617507761366367e-05, "epoch": 0.4255672845150029, "percentage": 21.28, "elapsed_time": "13:41:29", "remaining_time": "2 days, 2:39:00", "throughput": "0.00", "total_tokens": 0}
|
| 65 |
+
{"current_steps": 6500, "total_steps": 30076, "loss": 0.5633, "learning_rate": 9.594936690750129e-05, "epoch": 0.4322167733355498, "percentage": 21.61, "elapsed_time": "13:54:21", "remaining_time": "2 days, 2:26:18", "throughput": "0.00", "total_tokens": 0}
|
| 66 |
+
{"current_steps": 6600, "total_steps": 30076, "loss": 0.5269, "learning_rate": 9.571746660871058e-05, "epoch": 0.43886626215609675, "percentage": 21.94, "elapsed_time": "14:07:23", "remaining_time": "2 days, 2:14:07", "throughput": "0.00", "total_tokens": 0}
|
| 67 |
+
{"current_steps": 6700, "total_steps": 30076, "loss": 0.5105, "learning_rate": 9.547940795533627e-05, "epoch": 0.44551575097664364, "percentage": 22.28, "elapsed_time": "14:20:36", "remaining_time": "2 days, 2:02:36", "throughput": "0.00", "total_tokens": 0}
|
| 68 |
+
{"current_steps": 6800, "total_steps": 30076, "loss": 0.5301, "learning_rate": 9.523522301498202e-05, "epoch": 0.4521652397971906, "percentage": 22.61, "elapsed_time": "14:33:08", "remaining_time": "2 days, 1:48:43", "throughput": "0.00", "total_tokens": 0}
|
| 69 |
+
{"current_steps": 6900, "total_steps": 30076, "loss": 0.5173, "learning_rate": 9.498494468049072e-05, "epoch": 0.45881472861773753, "percentage": 22.94, "elapsed_time": "14:45:26", "remaining_time": "2 days, 1:34:02", "throughput": "0.00", "total_tokens": 0}
|
| 70 |
+
{"current_steps": 7000, "total_steps": 30076, "loss": 0.5284, "learning_rate": 9.472860666551369e-05, "epoch": 0.4654642174382844, "percentage": 23.27, "elapsed_time": "14:58:19", "remaining_time": "2 days, 1:21:22", "throughput": "0.00", "total_tokens": 0}
|
| 71 |
+
{"current_steps": 7100, "total_steps": 30076, "loss": 0.5317, "learning_rate": 9.446624349996929e-05, "epoch": 0.47211370625883137, "percentage": 23.61, "elapsed_time": "15:10:31", "remaining_time": "2 days, 1:06:32", "throughput": "0.00", "total_tokens": 0}
|
| 72 |
+
{"current_steps": 7200, "total_steps": 30076, "loss": 0.5195, "learning_rate": 9.419789052539157e-05, "epoch": 0.47876319507937826, "percentage": 23.94, "elapsed_time": "15:23:09", "remaining_time": "2 days, 0:53:05", "throughput": "0.00", "total_tokens": 0}
|
| 73 |
+
{"current_steps": 7300, "total_steps": 30076, "loss": 0.5274, "learning_rate": 9.392358389016961e-05, "epoch": 0.4854126838999252, "percentage": 24.27, "elapsed_time": "15:36:16", "remaining_time": "2 days, 0:41:10", "throughput": "0.00", "total_tokens": 0}
|
| 74 |
+
{"current_steps": 7400, "total_steps": 30076, "loss": 0.523, "learning_rate": 9.364336054467819e-05, "epoch": 0.4920621727204721, "percentage": 24.6, "elapsed_time": "15:49:30", "remaining_time": "2 days, 0:29:36", "throughput": "0.00", "total_tokens": 0}
|
| 75 |
+
{"current_steps": 7500, "total_steps": 30076, "loss": 0.5294, "learning_rate": 9.335725823630035e-05, "epoch": 0.49871166154101904, "percentage": 24.94, "elapsed_time": "16:02:46", "remaining_time": "2 days, 0:18:06", "throughput": "0.00", "total_tokens": 0}
|
| 76 |
+
{"current_steps": 7600, "total_steps": 30076, "loss": 0.5087, "learning_rate": 9.306531550434268e-05, "epoch": 0.5053611503615659, "percentage": 25.27, "elapsed_time": "16:15:24", "remaining_time": "2 days, 0:04:39", "throughput": "0.00", "total_tokens": 0}
|
| 77 |
+
{"current_steps": 7700, "total_steps": 30076, "loss": 0.5219, "learning_rate": 9.276757167484389e-05, "epoch": 0.5120106391821129, "percentage": 25.6, "elapsed_time": "16:28:56", "remaining_time": "1 day, 23:53:50", "throughput": "0.00", "total_tokens": 0}
|
| 78 |
+
{"current_steps": 7800, "total_steps": 30076, "loss": 0.5582, "learning_rate": 9.246406685527739e-05, "epoch": 0.5186601280026598, "percentage": 25.93, "elapsed_time": "16:41:30", "remaining_time": "1 day, 23:40:13", "throughput": "0.00", "total_tokens": 0}
|
| 79 |
+
{"current_steps": 7900, "total_steps": 30076, "loss": 0.5394, "learning_rate": 9.21548419291487e-05, "epoch": 0.5253096168232068, "percentage": 26.27, "elapsed_time": "16:53:52", "remaining_time": "1 day, 23:26:02", "throughput": "0.00", "total_tokens": 0}
|
| 80 |
+
{"current_steps": 8000, "total_steps": 30076, "loss": 0.5299, "learning_rate": 9.183993855048811e-05, "epoch": 0.5319591056437536, "percentage": 26.6, "elapsed_time": "17:06:12", "remaining_time": "1 day, 23:11:49", "throughput": "0.00", "total_tokens": 0}
|
| 81 |
+
{"current_steps": 8100, "total_steps": 30076, "loss": 0.5307, "learning_rate": 9.151939913823988e-05, "epoch": 0.5386085944643005, "percentage": 26.93, "elapsed_time": "17:19:22", "remaining_time": "1 day, 22:59:54", "throughput": "0.00", "total_tokens": 0}
|
| 82 |
+
{"current_steps": 8200, "total_steps": 30076, "loss": 0.5004, "learning_rate": 9.119326687054802e-05, "epoch": 0.5452580832848475, "percentage": 27.26, "elapsed_time": "17:32:01", "remaining_time": "1 day, 22:46:36", "throughput": "0.00", "total_tokens": 0}
|
| 83 |
+
{"current_steps": 8300, "total_steps": 30076, "loss": 0.5453, "learning_rate": 9.086158567894013e-05, "epoch": 0.5519075721053944, "percentage": 27.6, "elapsed_time": "17:44:55", "remaining_time": "1 day, 22:33:57", "throughput": "0.00", "total_tokens": 0}
|
| 84 |
+
{"current_steps": 8400, "total_steps": 30076, "loss": 0.5003, "learning_rate": 9.052440024240956e-05, "epoch": 0.5585570609259414, "percentage": 27.93, "elapsed_time": "17:57:23", "remaining_time": "1 day, 22:20:10", "throughput": "0.00", "total_tokens": 0}
|
| 85 |
+
{"current_steps": 8500, "total_steps": 30076, "loss": 0.508, "learning_rate": 9.018175598139696e-05, "epoch": 0.5652065497464882, "percentage": 28.26, "elapsed_time": "18:09:46", "remaining_time": "1 day, 22:06:14", "throughput": "0.00", "total_tokens": 0}
|
| 86 |
+
{"current_steps": 8600, "total_steps": 30076, "loss": 0.4788, "learning_rate": 8.983369905167191e-05, "epoch": 0.5718560385670352, "percentage": 28.59, "elapsed_time": "18:22:54", "remaining_time": "1 day, 21:54:12", "throughput": "0.00", "total_tokens": 0}
|
| 87 |
+
{"current_steps": 8700, "total_steps": 30076, "loss": 0.4994, "learning_rate": 8.948027633811557e-05, "epoch": 0.5785055273875821, "percentage": 28.93, "elapsed_time": "18:35:59", "remaining_time": "1 day, 21:41:59", "throughput": "0.00", "total_tokens": 0}
|
| 88 |
+
{"current_steps": 8800, "total_steps": 30076, "loss": 0.546, "learning_rate": 8.912153544840507e-05, "epoch": 0.585155016208129, "percentage": 29.26, "elapsed_time": "18:49:02", "remaining_time": "1 day, 21:29:41", "throughput": "0.00", "total_tokens": 0}
|
| 89 |
+
{"current_steps": 8900, "total_steps": 30076, "loss": 0.4939, "learning_rate": 8.875752470660043e-05, "epoch": 0.5918045050286759, "percentage": 29.59, "elapsed_time": "19:02:03", "remaining_time": "1 day, 21:17:20", "throughput": "0.00", "total_tokens": 0}
|
| 90 |
+
{"current_steps": 9000, "total_steps": 30076, "loss": 0.5288, "learning_rate": 8.838829314663522e-05, "epoch": 0.5984539938492228, "percentage": 29.92, "elapsed_time": "19:15:50", "remaining_time": "1 day, 21:06:43", "throughput": "0.00", "total_tokens": 0}
|
| 91 |
+
{"current_steps": 9100, "total_steps": 30076, "loss": 0.5176, "learning_rate": 8.80138905057114e-05, "epoch": 0.6051034826697698, "percentage": 30.26, "elapsed_time": "19:28:53", "remaining_time": "1 day, 20:54:21", "throughput": "0.00", "total_tokens": 0}
|
| 92 |
+
{"current_steps": 9200, "total_steps": 30076, "loss": 0.4898, "learning_rate": 8.763436721759952e-05, "epoch": 0.6117529714903167, "percentage": 30.59, "elapsed_time": "19:41:06", "remaining_time": "1 day, 20:40:05", "throughput": "0.00", "total_tokens": 0}
|
| 93 |
+
{"current_steps": 9300, "total_steps": 30076, "loss": 0.5011, "learning_rate": 8.724977440584497e-05, "epoch": 0.6184024603108637, "percentage": 30.92, "elapsed_time": "19:53:46", "remaining_time": "1 day, 20:26:52", "throughput": "0.00", "total_tokens": 0}
|
| 94 |
+
{"current_steps": 9400, "total_steps": 30076, "loss": 0.5101, "learning_rate": 8.686016387688153e-05, "epoch": 0.6250519491314105, "percentage": 31.25, "elapsed_time": "20:07:13", "remaining_time": "1 day, 20:15:23", "throughput": "0.00", "total_tokens": 0}
|
| 95 |
+
{"current_steps": 9500, "total_steps": 30076, "loss": 0.5023, "learning_rate": 8.646558811305268e-05, "epoch": 0.6317014379519574, "percentage": 31.59, "elapsed_time": "20:20:11", "remaining_time": "1 day, 20:02:48", "throughput": "0.00", "total_tokens": 0}
|
| 96 |
+
{"current_steps": 9600, "total_steps": 30076, "loss": 0.5055, "learning_rate": 8.60661002655421e-05, "epoch": 0.6383509267725044, "percentage": 31.92, "elapsed_time": "20:33:41", "remaining_time": "1 day, 19:51:21", "throughput": "0.00", "total_tokens": 0}
|
| 97 |
+
{"current_steps": 9700, "total_steps": 30076, "loss": 0.4997, "learning_rate": 8.566175414721384e-05, "epoch": 0.6450004155930513, "percentage": 32.25, "elapsed_time": "20:47:11", "remaining_time": "1 day, 19:39:52", "throughput": "0.00", "total_tokens": 0}
|
| 98 |
+
{"current_steps": 9800, "total_steps": 30076, "loss": 0.5104, "learning_rate": 8.525260422536358e-05, "epoch": 0.6516499044135982, "percentage": 32.58, "elapsed_time": "20:59:39", "remaining_time": "1 day, 19:26:11", "throughput": "0.00", "total_tokens": 0}
|
| 99 |
+
{"current_steps": 9900, "total_steps": 30076, "loss": 0.5281, "learning_rate": 8.483870561438161e-05, "epoch": 0.6582993932341451, "percentage": 32.92, "elapsed_time": "21:12:36", "remaining_time": "1 day, 19:13:33", "throughput": "0.00", "total_tokens": 0}
|
| 100 |
+
{"current_steps": 10000, "total_steps": 30076, "loss": 0.5147, "learning_rate": 8.442011406832859e-05, "epoch": 0.664948882054692, "percentage": 33.25, "elapsed_time": "21:25:14", "remaining_time": "1 day, 19:00:15", "throughput": "0.00", "total_tokens": 0}
|
| 101 |
+
{"current_steps": 10000, "total_steps": 30076, "eval_loss": 0.4313640594482422, "epoch": 0.664948882054692, "percentage": 33.25, "elapsed_time": "22:37:35", "remaining_time": "1 day, 21:25:29", "throughput": "0.00", "total_tokens": 0}
|
| 102 |
+
{"current_steps": 10100, "total_steps": 30076, "loss": 0.5229, "learning_rate": 8.399688597342535e-05, "epoch": 0.671598370875239, "percentage": 33.58, "elapsed_time": "22:51:00", "remaining_time": "1 day, 21:11:37", "throughput": "0.00", "total_tokens": 0}
|
| 103 |
+
{"current_steps": 10200, "total_steps": 30076, "loss": 0.5008, "learning_rate": 8.356907834045726e-05, "epoch": 0.6782478596957859, "percentage": 33.91, "elapsed_time": "23:03:58", "remaining_time": "1 day, 20:56:50", "throughput": "0.00", "total_tokens": 0}
|
| 104 |
+
{"current_steps": 10300, "total_steps": 30076, "loss": 0.5084, "learning_rate": 8.313674879709475e-05, "epoch": 0.6848973485163328, "percentage": 34.25, "elapsed_time": "23:17:34", "remaining_time": "1 day, 20:43:21", "throughput": "0.00", "total_tokens": 0}
|
| 105 |
+
{"current_steps": 10400, "total_steps": 30076, "loss": 0.5049, "learning_rate": 8.269995558013049e-05, "epoch": 0.6915468373368797, "percentage": 34.58, "elapsed_time": "23:29:46", "remaining_time": "1 day, 20:27:11", "throughput": "0.00", "total_tokens": 0}
|
| 106 |
+
{"current_steps": 10500, "total_steps": 30076, "loss": 0.4711, "learning_rate": 8.225875752763468e-05, "epoch": 0.6981963261574267, "percentage": 34.91, "elapsed_time": "23:42:29", "remaining_time": "1 day, 20:12:03", "throughput": "0.00", "total_tokens": 0}
|
| 107 |
+
{"current_steps": 10600, "total_steps": 30076, "loss": 0.4835, "learning_rate": 8.181321407102929e-05, "epoch": 0.7048458149779736, "percentage": 35.24, "elapsed_time": "23:55:37", "remaining_time": "1 day, 19:57:45", "throughput": "0.00", "total_tokens": 0}
|
| 108 |
+
{"current_steps": 10700, "total_steps": 30076, "loss": 0.4849, "learning_rate": 8.136338522708233e-05, "epoch": 0.7114953037985204, "percentage": 35.58, "elapsed_time": "1 day, 0:09:09", "remaining_time": "1 day, 19:44:11", "throughput": "0.00", "total_tokens": 0}
|
| 109 |
+
{"current_steps": 10800, "total_steps": 30076, "loss": 0.4881, "learning_rate": 8.090933158982338e-05, "epoch": 0.7181447926190674, "percentage": 35.91, "elapsed_time": "1 day, 0:22:20", "remaining_time": "1 day, 19:30:00", "throughput": "0.00", "total_tokens": 0}
|
| 110 |
+
{"current_steps": 10900, "total_steps": 30076, "loss": 0.4699, "learning_rate": 8.045111432238121e-05, "epoch": 0.7247942814396143, "percentage": 36.24, "elapsed_time": "1 day, 0:34:26", "remaining_time": "1 day, 19:13:56", "throughput": "0.00", "total_tokens": 0}
|
| 111 |
+
{"current_steps": 11000, "total_steps": 30076, "loss": 0.5022, "learning_rate": 7.998879514874491e-05, "epoch": 0.7314437702601613, "percentage": 36.57, "elapsed_time": "1 day, 0:47:56", "remaining_time": "1 day, 19:00:21", "throughput": "0.00", "total_tokens": 0}
|
| 112 |
+
{"current_steps": 11100, "total_steps": 30076, "loss": 0.4933, "learning_rate": 7.952243634544929e-05, "epoch": 0.7380932590807082, "percentage": 36.91, "elapsed_time": "1 day, 1:01:37", "remaining_time": "1 day, 18:47:05", "throughput": "0.00", "total_tokens": 0}
|
| 113 |
+
{"current_steps": 11200, "total_steps": 30076, "loss": 0.4677, "learning_rate": 7.905210073318605e-05, "epoch": 0.744742747901255, "percentage": 37.24, "elapsed_time": "1 day, 1:14:27", "remaining_time": "1 day, 18:32:24", "throughput": "0.00", "total_tokens": 0}
|
| 114 |
+
{"current_steps": 11300, "total_steps": 30076, "loss": 0.4757, "learning_rate": 7.857785166834144e-05, "epoch": 0.751392236721802, "percentage": 37.57, "elapsed_time": "1 day, 1:27:42", "remaining_time": "1 day, 18:18:26", "throughput": "0.00", "total_tokens": 0}
|
| 115 |
+
{"current_steps": 11400, "total_steps": 30076, "loss": 0.4938, "learning_rate": 7.809975303446195e-05, "epoch": 0.7580417255423489, "percentage": 37.9, "elapsed_time": "1 day, 1:41:14", "remaining_time": "1 day, 18:04:56", "throughput": "0.00", "total_tokens": 0}
|
| 116 |
+
{"current_steps": 11500, "total_steps": 30076, "loss": 0.4733, "learning_rate": 7.761786923364878e-05, "epoch": 0.7646912143628959, "percentage": 38.24, "elapsed_time": "1 day, 1:54:28", "remaining_time": "1 day, 17:50:56", "throughput": "0.00", "total_tokens": 0}
|
| 117 |
+
{"current_steps": 11600, "total_steps": 30076, "loss": 0.4655, "learning_rate": 7.713226517788275e-05, "epoch": 0.7713407031834427, "percentage": 38.57, "elapsed_time": "1 day, 2:07:34", "remaining_time": "1 day, 17:36:46", "throughput": "0.00", "total_tokens": 0}
|
| 118 |
+
{"current_steps": 11700, "total_steps": 30076, "loss": 0.4933, "learning_rate": 7.664300628028017e-05, "epoch": 0.7779901920039897, "percentage": 38.9, "elapsed_time": "1 day, 2:20:34", "remaining_time": "1 day, 17:22:26", "throughput": "0.00", "total_tokens": 0}
|
| 119 |
+
{"current_steps": 11800, "total_steps": 30076, "loss": 0.4699, "learning_rate": 7.615015844628157e-05, "epoch": 0.7846396808245366, "percentage": 39.23, "elapsed_time": "1 day, 2:33:12", "remaining_time": "1 day, 17:07:34", "throughput": "0.00", "total_tokens": 0}
|
| 120 |
+
{"current_steps": 11900, "total_steps": 30076, "loss": 0.4514, "learning_rate": 7.565378806477377e-05, "epoch": 0.7912891696450836, "percentage": 39.57, "elapsed_time": "1 day, 2:46:16", "remaining_time": "1 day, 16:53:24", "throughput": "0.00", "total_tokens": 0}
|
| 121 |
+
{"current_steps": 12000, "total_steps": 30076, "loss": 0.4532, "learning_rate": 7.515396199914708e-05, "epoch": 0.7979386584656305, "percentage": 39.9, "elapsed_time": "1 day, 2:59:45", "remaining_time": "1 day, 16:39:54", "throughput": "0.00", "total_tokens": 0}
|
| 122 |
+
{"current_steps": 12100, "total_steps": 30076, "loss": 0.4749, "learning_rate": 7.46507475782885e-05, "epoch": 0.8045881472861773, "percentage": 40.23, "elapsed_time": "1 day, 3:12:48", "remaining_time": "1 day, 16:25:44", "throughput": "0.00", "total_tokens": 0}
|
| 123 |
+
{"current_steps": 12200, "total_steps": 30076, "loss": 0.4977, "learning_rate": 7.414421258751212e-05, "epoch": 0.8112376361067243, "percentage": 40.56, "elapsed_time": "1 day, 3:25:35", "remaining_time": "1 day, 16:11:12", "throughput": "0.00", "total_tokens": 0}
|
| 124 |
+
{"current_steps": 12300, "total_steps": 30076, "loss": 0.4544, "learning_rate": 7.363442525942826e-05, "epoch": 0.8178871249272712, "percentage": 40.9, "elapsed_time": "1 day, 3:39:14", "remaining_time": "1 day, 15:57:56", "throughput": "0.00", "total_tokens": 0}
|
| 125 |
+
{"current_steps": 12400, "total_steps": 30076, "loss": 0.5053, "learning_rate": 7.31214542647521e-05, "epoch": 0.8245366137478182, "percentage": 41.23, "elapsed_time": "1 day, 3:52:47", "remaining_time": "1 day, 15:44:31", "throughput": "0.00", "total_tokens": 0}
|
| 126 |
+
{"current_steps": 12500, "total_steps": 30076, "loss": 0.4648, "learning_rate": 7.260536870305347e-05, "epoch": 0.831186102568365, "percentage": 41.56, "elapsed_time": "1 day, 4:05:46", "remaining_time": "1 day, 15:30:19", "throughput": "0.00", "total_tokens": 0}
|
| 127 |
+
{"current_steps": 12600, "total_steps": 30076, "loss": 0.4313, "learning_rate": 7.208623809344879e-05, "epoch": 0.837835591388912, "percentage": 41.89, "elapsed_time": "1 day, 4:18:36", "remaining_time": "1 day, 15:15:56", "throughput": "0.00", "total_tokens": 0}
|
| 128 |
+
{"current_steps": 12700, "total_steps": 30076, "loss": 0.4816, "learning_rate": 7.156413236523656e-05, "epoch": 0.8444850802094589, "percentage": 42.23, "elapsed_time": "1 day, 4:31:08", "remaining_time": "1 day, 15:01:09", "throughput": "0.00", "total_tokens": 0}
|
| 129 |
+
{"current_steps": 12800, "total_steps": 30076, "loss": 0.4534, "learning_rate": 7.103912184847757e-05, "epoch": 0.8511345690300058, "percentage": 42.56, "elapsed_time": "1 day, 4:43:59", "remaining_time": "1 day, 14:46:51", "throughput": "0.00", "total_tokens": 0}
|
| 130 |
+
{"current_steps": 12900, "total_steps": 30076, "loss": 0.4304, "learning_rate": 7.051127726452102e-05, "epoch": 0.8577840578505528, "percentage": 42.89, "elapsed_time": "1 day, 4:56:09", "remaining_time": "1 day, 14:31:38", "throughput": "0.00", "total_tokens": 0}
|
| 131 |
+
{"current_steps": 13000, "total_steps": 30076, "loss": 0.4511, "learning_rate": 6.998066971647817e-05, "epoch": 0.8644335466710996, "percentage": 43.22, "elapsed_time": "1 day, 5:08:44", "remaining_time": "1 day, 14:17:02", "throughput": "0.00", "total_tokens": 0}
|
| 132 |
+
{"current_steps": 13100, "total_steps": 30076, "loss": 0.4412, "learning_rate": 6.944737067964429e-05, "epoch": 0.8710830354916466, "percentage": 43.56, "elapsed_time": "1 day, 5:21:49", "remaining_time": "1 day, 14:03:06", "throughput": "0.00", "total_tokens": 0}
|
| 133 |
+
{"current_steps": 13200, "total_steps": 30076, "loss": 0.4455, "learning_rate": 6.891145199187065e-05, "epoch": 0.8777325243121935, "percentage": 43.89, "elapsed_time": "1 day, 5:34:39", "remaining_time": "1 day, 13:48:52", "throughput": "0.00", "total_tokens": 0}
|
| 134 |
+
{"current_steps": 13300, "total_steps": 30076, "loss": 0.4598, "learning_rate": 6.837298584388771e-05, "epoch": 0.8843820131327405, "percentage": 44.22, "elapsed_time": "1 day, 5:48:05", "remaining_time": "1 day, 13:35:24", "throughput": "0.00", "total_tokens": 0}
|
| 135 |
+
{"current_steps": 13400, "total_steps": 30076, "loss": 0.4561, "learning_rate": 6.783204476958058e-05, "epoch": 0.8910315019532873, "percentage": 44.55, "elapsed_time": "1 day, 6:00:52", "remaining_time": "1 day, 13:21:08", "throughput": "0.00", "total_tokens": 0}
|
| 136 |
+
{"current_steps": 13500, "total_steps": 30076, "loss": 0.4818, "learning_rate": 6.728870163621836e-05, "epoch": 0.8976809907738342, "percentage": 44.89, "elapsed_time": "1 day, 6:13:50", "remaining_time": "1 day, 13:07:07", "throughput": "0.00", "total_tokens": 0}
|
| 137 |
+
{"current_steps": 13600, "total_steps": 30076, "loss": 0.4322, "learning_rate": 6.674302963463876e-05, "epoch": 0.9043304795943812, "percentage": 45.22, "elapsed_time": "1 day, 6:27:00", "remaining_time": "1 day, 12:53:22", "throughput": "0.00", "total_tokens": 0}
|
| 138 |
+
{"current_steps": 13700, "total_steps": 30076, "loss": 0.4601, "learning_rate": 6.61951022693887e-05, "epoch": 0.9109799684149281, "percentage": 45.55, "elapsed_time": "1 day, 6:38:58", "remaining_time": "1 day, 12:38:11", "throughput": "0.00", "total_tokens": 0}
|
| 139 |
+
{"current_steps": 13800, "total_steps": 30076, "loss": 0.4787, "learning_rate": 6.564499334882312e-05, "epoch": 0.9176294572354751, "percentage": 45.88, "elapsed_time": "1 day, 6:51:55", "remaining_time": "1 day, 12:24:11", "throughput": "0.00", "total_tokens": 0}
|
| 140 |
+
{"current_steps": 13900, "total_steps": 30076, "loss": 0.4766, "learning_rate": 6.509277697516255e-05, "epoch": 0.9242789460560219, "percentage": 46.22, "elapsed_time": "1 day, 7:04:19", "remaining_time": "1 day, 12:09:35", "throughput": "0.00", "total_tokens": 0}
|
| 141 |
+
{"current_steps": 14000, "total_steps": 30076, "loss": 0.4614, "learning_rate": 6.453852753451119e-05, "epoch": 0.9309284348765688, "percentage": 46.55, "elapsed_time": "1 day, 7:17:27", "remaining_time": "1 day, 11:55:51", "throughput": "0.00", "total_tokens": 0}
|
| 142 |
+
{"current_steps": 14100, "total_steps": 30076, "loss": 0.48, "learning_rate": 6.39823196868368e-05, "epoch": 0.9375779236971158, "percentage": 46.88, "elapsed_time": "1 day, 7:30:39", "remaining_time": "1 day, 11:42:12", "throughput": "0.00", "total_tokens": 0}
|
| 143 |
+
{"current_steps": 14200, "total_steps": 30076, "loss": 0.4717, "learning_rate": 6.342422835591368e-05, "epoch": 0.9442274125176627, "percentage": 47.21, "elapsed_time": "1 day, 7:44:02", "remaining_time": "1 day, 11:28:46", "throughput": "0.00", "total_tokens": 0}
|
| 144 |
+
{"current_steps": 14300, "total_steps": 30076, "loss": 0.4609, "learning_rate": 6.286432871923e-05, "epoch": 0.9508769013382097, "percentage": 47.55, "elapsed_time": "1 day, 7:57:40", "remaining_time": "1 day, 11:15:36", "throughput": "0.00", "total_tokens": 0}
|
| 145 |
+
{"current_steps": 14400, "total_steps": 30076, "loss": 0.4233, "learning_rate": 6.230269619786111e-05, "epoch": 0.9575263901587565, "percentage": 47.88, "elapsed_time": "1 day, 8:10:22", "remaining_time": "1 day, 11:01:25", "throughput": "0.00", "total_tokens": 0}
|
| 146 |
+
{"current_steps": 14500, "total_steps": 30076, "loss": 0.4368, "learning_rate": 6.173940644630996e-05, "epoch": 0.9641758789793035, "percentage": 48.21, "elapsed_time": "1 day, 8:23:23", "remaining_time": "1 day, 10:47:36", "throughput": "0.00", "total_tokens": 0}
|
| 147 |
+
{"current_steps": 14600, "total_steps": 30076, "loss": 0.4205, "learning_rate": 6.117453534231606e-05, "epoch": 0.9708253677998504, "percentage": 48.54, "elapsed_time": "1 day, 8:36:29", "remaining_time": "1 day, 10:33:53", "throughput": "0.00", "total_tokens": 0}
|
| 148 |
+
{"current_steps": 14700, "total_steps": 30076, "loss": 0.4247, "learning_rate": 6.060815897663447e-05, "epoch": 0.9774748566203973, "percentage": 48.88, "elapsed_time": "1 day, 8:49:43", "remaining_time": "1 day, 10:20:18", "throughput": "0.00", "total_tokens": 0}
|
| 149 |
+
{"current_steps": 14800, "total_steps": 30076, "loss": 0.4338, "learning_rate": 6.004035364278593e-05, "epoch": 0.9841243454409442, "percentage": 49.21, "elapsed_time": "1 day, 9:02:53", "remaining_time": "1 day, 10:06:39", "throughput": "0.00", "total_tokens": 0}
|
| 150 |
+
{"current_steps": 14900, "total_steps": 30076, "loss": 0.4114, "learning_rate": 5.9471195826779834e-05, "epoch": 0.9907738342614911, "percentage": 49.54, "elapsed_time": "1 day, 9:15:35", "remaining_time": "1 day, 9:52:33", "throughput": "0.00", "total_tokens": 0}
|
| 151 |
+
{"current_steps": 15000, "total_steps": 30076, "loss": 0.4209, "learning_rate": 5.8900762196811175e-05, "epoch": 0.9974233230820381, "percentage": 49.87, "elapsed_time": "1 day, 9:28:12", "remaining_time": "1 day, 9:38:23", "throughput": "0.00", "total_tokens": 0}
|
| 152 |
+
{"current_steps": 15100, "total_steps": 30076, "loss": 0.3562, "learning_rate": 5.832912959293304e-05, "epoch": 1.004072811902585, "percentage": 50.21, "elapsed_time": "1 day, 9:41:19", "remaining_time": "1 day, 9:24:43", "throughput": "0.00", "total_tokens": 0}
|
| 153 |
+
{"current_steps": 15200, "total_steps": 30076, "loss": 0.3388, "learning_rate": 5.775637501670579e-05, "epoch": 1.0107223007231319, "percentage": 50.54, "elapsed_time": "1 day, 9:54:11", "remaining_time": "1 day, 9:10:49", "throughput": "0.00", "total_tokens": 0}
|
| 154 |
+
{"current_steps": 15300, "total_steps": 30076, "loss": 0.3572, "learning_rate": 5.718257562082471e-05, "epoch": 1.017371789543679, "percentage": 50.87, "elapsed_time": "1 day, 10:06:58", "remaining_time": "1 day, 8:56:52", "throughput": "0.00", "total_tokens": 0}
|
| 155 |
+
{"current_steps": 15400, "total_steps": 30076, "loss": 0.3377, "learning_rate": 5.660780869872711e-05, "epoch": 1.0240212783642257, "percentage": 51.2, "elapsed_time": "1 day, 10:19:50", "remaining_time": "1 day, 8:43:00", "throughput": "0.00", "total_tokens": 0}
|
| 156 |
+
{"current_steps": 15500, "total_steps": 30076, "loss": 0.3561, "learning_rate": 5.6032151674180575e-05, "epoch": 1.0306707671847726, "percentage": 51.54, "elapsed_time": "1 day, 10:32:27", "remaining_time": "1 day, 8:28:55", "throughput": "0.00", "total_tokens": 0}
|
| 157 |
+
{"current_steps": 15600, "total_steps": 30076, "loss": 0.3698, "learning_rate": 5.5455682090853624e-05, "epoch": 1.0373202560053196, "percentage": 51.87, "elapsed_time": "1 day, 10:44:46", "remaining_time": "1 day, 8:14:33", "throughput": "0.00", "total_tokens": 0}
|
| 158 |
+
{"current_steps": 15700, "total_steps": 30076, "loss": 0.3223, "learning_rate": 5.4878477601870194e-05, "epoch": 1.0439697448258665, "percentage": 52.2, "elapsed_time": "1 day, 10:57:56", "remaining_time": "1 day, 8:01:00", "throughput": "0.00", "total_tokens": 0}
|
| 159 |
+
{"current_steps": 15800, "total_steps": 30076, "loss": 0.3483, "learning_rate": 5.430061595934941e-05, "epoch": 1.0506192336464135, "percentage": 52.53, "elapsed_time": "1 day, 11:10:08", "remaining_time": "1 day, 7:46:36", "throughput": "0.00", "total_tokens": 0}
|
| 160 |
+
{"current_steps": 15900, "total_steps": 30076, "loss": 0.3642, "learning_rate": 5.372217500393205e-05, "epoch": 1.0572687224669604, "percentage": 52.87, "elapsed_time": "1 day, 11:23:04", "remaining_time": "1 day, 7:32:52", "throughput": "0.00", "total_tokens": 0}
|
| 161 |
+
{"current_steps": 16000, "total_steps": 30076, "loss": 0.3234, "learning_rate": 5.314323265429501e-05, "epoch": 1.0639182112875072, "percentage": 53.2, "elapsed_time": "1 day, 11:35:47", "remaining_time": "1 day, 7:18:57", "throughput": "0.00", "total_tokens": 0}
|
| 162 |
+
{"current_steps": 16100, "total_steps": 30076, "loss": 0.3448, "learning_rate": 5.2563866896655275e-05, "epoch": 1.0705677001080542, "percentage": 53.53, "elapsed_time": "1 day, 11:48:28", "remaining_time": "1 day, 7:05:02", "throughput": "0.00", "total_tokens": 0}
|
| 163 |
+
{"current_steps": 16200, "total_steps": 30076, "loss": 0.3621, "learning_rate": 5.198415577426493e-05, "epoch": 1.077217188928601, "percentage": 53.86, "elapsed_time": "1 day, 12:01:50", "remaining_time": "1 day, 6:51:42", "throughput": "0.00", "total_tokens": 0}
|
| 164 |
+
{"current_steps": 16300, "total_steps": 30076, "loss": 0.3532, "learning_rate": 5.140417737689822e-05, "epoch": 1.0838666777491481, "percentage": 54.2, "elapsed_time": "1 day, 12:15:11", "remaining_time": "1 day, 6:38:21", "throughput": "0.00", "total_tokens": 0}
|
| 165 |
+
{"current_steps": 16400, "total_steps": 30076, "loss": 0.3424, "learning_rate": 5.0824009830332606e-05, "epoch": 1.090516166569695, "percentage": 54.53, "elapsed_time": "1 day, 12:28:08", "remaining_time": "1 day, 6:24:41", "throughput": "0.00", "total_tokens": 0}
|
| 166 |
+
{"current_steps": 16500, "total_steps": 30076, "loss": 0.3465, "learning_rate": 5.02437312858248e-05, "epoch": 1.0971656553902418, "percentage": 54.86, "elapsed_time": "1 day, 12:41:41", "remaining_time": "1 day, 6:11:31", "throughput": "0.00", "total_tokens": 0}
|
| 167 |
+
{"current_steps": 16600, "total_steps": 30076, "loss": 0.3471, "learning_rate": 4.966341990958347e-05, "epoch": 1.1038151442107889, "percentage": 55.19, "elapsed_time": "1 day, 12:54:09", "remaining_time": "1 day, 5:57:28", "throughput": "0.00", "total_tokens": 0}
|
| 168 |
+
{"current_steps": 16700, "total_steps": 30076, "loss": 0.3408, "learning_rate": 4.908315387223985e-05, "epoch": 1.1104646330313357, "percentage": 55.53, "elapsed_time": "1 day, 13:06:48", "remaining_time": "1 day, 5:43:34", "throughput": "0.00", "total_tokens": 0}
|
| 169 |
+
{"current_steps": 16800, "total_steps": 30076, "loss": 0.3351, "learning_rate": 4.850301133831786e-05, "epoch": 1.1171141218518827, "percentage": 55.86, "elapsed_time": "1 day, 13:20:01", "remaining_time": "1 day, 5:30:08", "throughput": "0.00", "total_tokens": 0}
|
| 170 |
+
{"current_steps": 16900, "total_steps": 30076, "loss": 0.3531, "learning_rate": 4.792307045570486e-05, "epoch": 1.1237636106724296, "percentage": 56.19, "elapsed_time": "1 day, 13:33:18", "remaining_time": "1 day, 5:16:46", "throughput": "0.00", "total_tokens": 0}
|
| 171 |
+
{"current_steps": 17000, "total_steps": 30076, "loss": 0.3171, "learning_rate": 4.734340934512492e-05, "epoch": 1.1304130994929764, "percentage": 56.52, "elapsed_time": "1 day, 13:46:15", "remaining_time": "1 day, 5:03:09", "throughput": "0.00", "total_tokens": 0}
|
| 172 |
+
{"current_steps": 17100, "total_steps": 30076, "loss": 0.3388, "learning_rate": 4.6764106089615454e-05, "epoch": 1.1370625883135235, "percentage": 56.86, "elapsed_time": "1 day, 13:59:08", "remaining_time": "1 day, 4:49:28", "throughput": "0.00", "total_tokens": 0}
|
| 173 |
+
{"current_steps": 17200, "total_steps": 30076, "loss": 0.3356, "learning_rate": 4.618523872400921e-05, "epoch": 1.1437120771340703, "percentage": 57.19, "elapsed_time": "1 day, 14:12:00", "remaining_time": "1 day, 4:35:48", "throughput": "0.00", "total_tokens": 0}
|
| 174 |
+
{"current_steps": 17300, "total_steps": 30076, "loss": 0.3483, "learning_rate": 4.56068852244225e-05, "epoch": 1.1503615659546171, "percentage": 57.52, "elapsed_time": "1 day, 14:24:58", "remaining_time": "1 day, 4:22:13", "throughput": "0.00", "total_tokens": 0}
|
| 175 |
+
{"current_steps": 17400, "total_steps": 30076, "loss": 0.3223, "learning_rate": 4.5029123497751514e-05, "epoch": 1.1570110547751642, "percentage": 57.85, "elapsed_time": "1 day, 14:38:04", "remaining_time": "1 day, 4:08:43", "throughput": "0.00", "total_tokens": 0}
|
| 176 |
+
{"current_steps": 17500, "total_steps": 30076, "loss": 0.3181, "learning_rate": 4.445203137117788e-05, "epoch": 1.163660543595711, "percentage": 58.19, "elapsed_time": "1 day, 14:51:21", "remaining_time": "1 day, 3:55:22", "throughput": "0.00", "total_tokens": 0}
|
| 177 |
+
{"current_steps": 17600, "total_steps": 30076, "loss": 0.3416, "learning_rate": 4.3875686581685e-05, "epoch": 1.170310032416258, "percentage": 58.52, "elapsed_time": "1 day, 15:04:20", "remaining_time": "1 day, 3:41:49", "throughput": "0.00", "total_tokens": 0}
|
| 178 |
+
{"current_steps": 17700, "total_steps": 30076, "loss": 0.3269, "learning_rate": 4.330016676558651e-05, "epoch": 1.176959521236805, "percentage": 58.85, "elapsed_time": "1 day, 15:17:14", "remaining_time": "1 day, 3:28:12", "throughput": "0.00", "total_tokens": 0}
|
| 179 |
+
{"current_steps": 17800, "total_steps": 30076, "loss": 0.3166, "learning_rate": 4.272554944806831e-05, "epoch": 1.1836090100573518, "percentage": 59.18, "elapsed_time": "1 day, 15:30:34", "remaining_time": "1 day, 3:14:54", "throughput": "0.00", "total_tokens": 0}
|
| 180 |
+
{"current_steps": 17900, "total_steps": 30076, "loss": 0.3354, "learning_rate": 4.2151912032745547e-05, "epoch": 1.1902584988778988, "percentage": 59.52, "elapsed_time": "1 day, 15:43:15", "remaining_time": "1 day, 3:01:08", "throughput": "0.00", "total_tokens": 0}
|
| 181 |
+
{"current_steps": 18000, "total_steps": 30076, "loss": 0.3066, "learning_rate": 4.157933179123599e-05, "epoch": 1.1969079876984456, "percentage": 59.85, "elapsed_time": "1 day, 15:55:43", "remaining_time": "1 day, 2:47:16", "throughput": "0.00", "total_tokens": 0}
|
| 182 |
+
{"current_steps": 18100, "total_steps": 30076, "loss": 0.3427, "learning_rate": 4.100788585275125e-05, "epoch": 1.2035574765189927, "percentage": 60.18, "elapsed_time": "1 day, 16:08:26", "remaining_time": "1 day, 2:33:33", "throughput": "0.00", "total_tokens": 0}
|
| 183 |
+
{"current_steps": 18200, "total_steps": 30076, "loss": 0.3074, "learning_rate": 4.043765119370699e-05, "epoch": 1.2102069653395395, "percentage": 60.51, "elapsed_time": "1 day, 16:20:48", "remaining_time": "1 day, 2:19:38", "throughput": "0.00", "total_tokens": 0}
|
| 184 |
+
{"current_steps": 18300, "total_steps": 30076, "loss": 0.3172, "learning_rate": 3.9868704627354e-05, "epoch": 1.2168564541600864, "percentage": 60.85, "elapsed_time": "1 day, 16:33:24", "remaining_time": "1 day, 2:05:53", "throughput": "0.00", "total_tokens": 0}
|
| 185 |
+
{"current_steps": 18400, "total_steps": 30076, "loss": 0.3088, "learning_rate": 3.930112279343094e-05, "epoch": 1.2235059429806334, "percentage": 61.18, "elapsed_time": "1 day, 16:45:51", "remaining_time": "1 day, 1:52:03", "throughput": "0.00", "total_tokens": 0}
|
| 186 |
+
{"current_steps": 18500, "total_steps": 30076, "loss": 0.3053, "learning_rate": 3.8734982147840756e-05, "epoch": 1.2301554318011803, "percentage": 61.51, "elapsed_time": "1 day, 16:58:43", "remaining_time": "1 day, 1:38:30", "throughput": "0.00", "total_tokens": 0}
|
| 187 |
+
{"current_steps": 18600, "total_steps": 30076, "loss": 0.3178, "learning_rate": 3.817035895235159e-05, "epoch": 1.236804920621727, "percentage": 61.84, "elapsed_time": "1 day, 17:11:16", "remaining_time": "1 day, 1:24:44", "throughput": "0.00", "total_tokens": 0}
|
| 188 |
+
{"current_steps": 18700, "total_steps": 30076, "loss": 0.2876, "learning_rate": 3.760732926432407e-05, "epoch": 1.2434544094422741, "percentage": 62.18, "elapsed_time": "1 day, 17:24:17", "remaining_time": "1 day, 1:11:18", "throughput": "0.00", "total_tokens": 0}
|
| 189 |
+
{"current_steps": 18800, "total_steps": 30076, "loss": 0.3264, "learning_rate": 3.704596892646593e-05, "epoch": 1.250103898262821, "percentage": 62.51, "elapsed_time": "1 day, 17:37:13", "remaining_time": "1 day, 0:57:48", "throughput": "0.00", "total_tokens": 0}
|
| 190 |
+
{"current_steps": 18900, "total_steps": 30076, "loss": 0.3151, "learning_rate": 3.648635355661577e-05, "epoch": 1.256753387083368, "percentage": 62.84, "elapsed_time": "1 day, 17:50:08", "remaining_time": "1 day, 0:44:18", "throughput": "0.00", "total_tokens": 0}
|
| 191 |
+
{"current_steps": 19000, "total_steps": 30076, "loss": 0.3086, "learning_rate": 3.5928558537556895e-05, "epoch": 1.2634028759039149, "percentage": 63.17, "elapsed_time": "1 day, 18:03:24", "remaining_time": "1 day, 0:31:00", "throughput": "0.00", "total_tokens": 0}
|
| 192 |
+
{"current_steps": 19100, "total_steps": 30076, "loss": 0.2933, "learning_rate": 3.537265900686286e-05, "epoch": 1.270052364724462, "percentage": 63.51, "elapsed_time": "1 day, 18:16:29", "remaining_time": "1 day, 0:17:37", "throughput": "0.00", "total_tokens": 0}
|
| 193 |
+
{"current_steps": 19200, "total_steps": 30076, "loss": 0.2867, "learning_rate": 3.4818729846776254e-05, "epoch": 1.2767018535450088, "percentage": 63.84, "elapsed_time": "1 day, 18:29:57", "remaining_time": "1 day, 0:04:26", "throughput": "0.00", "total_tokens": 0}
|
| 194 |
+
{"current_steps": 19300, "total_steps": 30076, "loss": 0.2956, "learning_rate": 3.426684567412153e-05, "epoch": 1.2833513423655556, "percentage": 64.17, "elapsed_time": "1 day, 18:43:36", "remaining_time": "23:51:22", "throughput": "0.00", "total_tokens": 0}
|
| 195 |
+
{"current_steps": 19400, "total_steps": 30076, "loss": 0.2944, "learning_rate": 3.371708083025392e-05, "epoch": 1.2900008311861026, "percentage": 64.5, "elapsed_time": "1 day, 18:56:38", "remaining_time": "23:37:56", "throughput": "0.00", "total_tokens": 0}
|
| 196 |
+
{"current_steps": 19500, "total_steps": 30076, "loss": 0.2868, "learning_rate": 3.316950937104518e-05, "epoch": 1.2966503200066495, "percentage": 64.84, "elapsed_time": "1 day, 19:09:56", "remaining_time": "23:24:40", "throughput": "0.00", "total_tokens": 0}
|
| 197 |
+
{"current_steps": 19600, "total_steps": 30076, "loss": 0.306, "learning_rate": 3.26242050569081e-05, "epoch": 1.3032998088271963, "percentage": 65.17, "elapsed_time": "1 day, 19:22:45", "remaining_time": "23:11:08", "throughput": "0.00", "total_tokens": 0}
|
| 198 |
+
{"current_steps": 19700, "total_steps": 30076, "loss": 0.294, "learning_rate": 3.208124134286038e-05, "epoch": 1.3099492976477434, "percentage": 65.5, "elapsed_time": "1 day, 19:34:59", "remaining_time": "22:57:19", "throughput": "0.00", "total_tokens": 0}
|
| 199 |
+
{"current_steps": 19800, "total_steps": 30076, "loss": 0.3237, "learning_rate": 3.1540691368630185e-05, "epoch": 1.3165987864682902, "percentage": 65.83, "elapsed_time": "1 day, 19:47:33", "remaining_time": "22:43:40", "throughput": "0.00", "total_tokens": 0}
|
| 200 |
+
{"current_steps": 19900, "total_steps": 30076, "loss": 0.3111, "learning_rate": 3.100262794880363e-05, "epoch": 1.323248275288837, "percentage": 66.17, "elapsed_time": "1 day, 20:00:28", "remaining_time": "22:30:13", "throughput": "0.00", "total_tokens": 0}
|
| 201 |
+
{"current_steps": 20000, "total_steps": 30076, "loss": 0.3106, "learning_rate": 3.0467123563016513e-05, "epoch": 1.329897764109384, "percentage": 66.5, "elapsed_time": "1 day, 20:12:43", "remaining_time": "22:16:26", "throughput": "0.00", "total_tokens": 0}
|
| 202 |
+
{"current_steps": 20000, "total_steps": 30076, "eval_loss": 0.31190410256385803, "epoch": 1.329897764109384, "percentage": 66.5, "elapsed_time": "1 day, 21:24:50", "remaining_time": "22:52:46", "throughput": "0.00", "total_tokens": 0}
|
| 203 |
+
{"current_steps": 20100, "total_steps": 30076, "loss": 0.2899, "learning_rate": 2.9934250346190818e-05, "epoch": 1.336547252929931, "percentage": 66.83, "elapsed_time": "1 day, 21:37:13", "remaining_time": "22:38:31", "throughput": "0.00", "total_tokens": 0}
|
| 204 |
+
{"current_steps": 20200, "total_steps": 30076, "loss": 0.284, "learning_rate": 2.9404080078817924e-05, "epoch": 1.343196741750478, "percentage": 67.16, "elapsed_time": "1 day, 21:50:30", "remaining_time": "22:24:45", "throughput": "0.00", "total_tokens": 0}
|
| 205 |
+
{"current_steps": 20300, "total_steps": 30076, "loss": 0.2886, "learning_rate": 2.8876684177289404e-05, "epoch": 1.3498462305710248, "percentage": 67.5, "elapsed_time": "1 day, 22:04:03", "remaining_time": "22:11:06", "throughput": "0.00", "total_tokens": 0}
|
| 206 |
+
{"current_steps": 20400, "total_steps": 30076, "loss": 0.2738, "learning_rate": 2.8352133684276853e-05, "epoch": 1.3564957193915719, "percentage": 67.83, "elapsed_time": "1 day, 22:16:17", "remaining_time": "21:56:50", "throughput": "0.00", "total_tokens": 0}
|
| 207 |
+
{"current_steps": 20500, "total_steps": 30076, "loss": 0.2981, "learning_rate": 2.7830499259162213e-05, "epoch": 1.3631452082121187, "percentage": 68.16, "elapsed_time": "1 day, 22:29:02", "remaining_time": "21:42:49", "throughput": "0.00", "total_tokens": 0}
|
| 208 |
+
{"current_steps": 20600, "total_steps": 30076, "loss": 0.3159, "learning_rate": 2.7311851168519496e-05, "epoch": 1.3697946970326655, "percentage": 68.49, "elapsed_time": "1 day, 22:42:12", "remaining_time": "21:29:00", "throughput": "0.00", "total_tokens": 0}
|
| 209 |
+
{"current_steps": 20700, "total_steps": 30076, "loss": 0.3019, "learning_rate": 2.6796259276649693e-05, "epoch": 1.3764441858532126, "percentage": 68.83, "elapsed_time": "1 day, 22:54:48", "remaining_time": "21:14:57", "throughput": "0.00", "total_tokens": 0}
|
| 210 |
+
{"current_steps": 20800, "total_steps": 30076, "loss": 0.3029, "learning_rate": 2.6283793036169603e-05, "epoch": 1.3830936746737594, "percentage": 69.16, "elapsed_time": "1 day, 23:07:38", "remaining_time": "21:01:01", "throughput": "0.00", "total_tokens": 0}
|
| 211 |
+
{"current_steps": 20900, "total_steps": 30076, "loss": 0.2799, "learning_rate": 2.5774521478656343e-05, "epoch": 1.3897431634943063, "percentage": 69.49, "elapsed_time": "1 day, 23:20:36", "remaining_time": "20:47:08", "throughput": "0.00", "total_tokens": 0}
|
| 212 |
+
{"current_steps": 21000, "total_steps": 30076, "loss": 0.2783, "learning_rate": 2.52685132053484e-05, "epoch": 1.3963926523148533, "percentage": 69.82, "elapsed_time": "1 day, 23:33:53", "remaining_time": "20:33:25", "throughput": "0.00", "total_tokens": 0}
|
| 213 |
+
{"current_steps": 21100, "total_steps": 30076, "loss": 0.3043, "learning_rate": 2.4765836377904787e-05, "epoch": 1.4030421411354002, "percentage": 70.16, "elapsed_time": "1 day, 23:47:05", "remaining_time": "20:19:40", "throughput": "0.00", "total_tokens": 0}
|
| 214 |
+
{"current_steps": 21200, "total_steps": 30076, "loss": 0.2823, "learning_rate": 2.4266558709223293e-05, "epoch": 1.4096916299559472, "percentage": 70.49, "elapsed_time": "2 days, 0:00:19", "remaining_time": "20:05:55", "throughput": "0.00", "total_tokens": 0}
|
| 215 |
+
{"current_steps": 21300, "total_steps": 30076, "loss": 0.2899, "learning_rate": 2.377074745431931e-05, "epoch": 1.416341118776494, "percentage": 70.82, "elapsed_time": "2 days, 0:13:26", "remaining_time": "19:52:09", "throughput": "0.00", "total_tokens": 0}
|
| 216 |
+
{"current_steps": 21400, "total_steps": 30076, "loss": 0.259, "learning_rate": 2.3278469401266178e-05, "epoch": 1.422990607597041, "percentage": 71.15, "elapsed_time": "2 days, 0:26:17", "remaining_time": "19:38:16", "throughput": "0.00", "total_tokens": 0}
|
| 217 |
+
{"current_steps": 21500, "total_steps": 30076, "loss": 0.2508, "learning_rate": 2.2789790862198628e-05, "epoch": 1.429640096417588, "percentage": 71.49, "elapsed_time": "2 days, 0:38:56", "remaining_time": "19:24:19", "throughput": "0.00", "total_tokens": 0}
|
| 218 |
+
{"current_steps": 21600, "total_steps": 30076, "loss": 0.2701, "learning_rate": 2.2304777664380176e-05, "epoch": 1.4362895852381348, "percentage": 71.82, "elapsed_time": "2 days, 0:51:39", "remaining_time": "19:10:24", "throughput": "0.00", "total_tokens": 0}
|
| 219 |
+
{"current_steps": 21700, "total_steps": 30076, "loss": 0.2741, "learning_rate": 2.182349514133583e-05, "epoch": 1.4429390740586818, "percentage": 72.15, "elapsed_time": "2 days, 1:04:16", "remaining_time": "18:56:27", "throughput": "0.00", "total_tokens": 0}
|
| 220 |
+
{"current_steps": 21800, "total_steps": 30076, "loss": 0.2608, "learning_rate": 2.134600812405151e-05, "epoch": 1.4495885628792287, "percentage": 72.48, "elapsed_time": "2 days, 1:17:01", "remaining_time": "18:42:35", "throughput": "0.00", "total_tokens": 0}
|
| 221 |
+
{"current_steps": 21900, "total_steps": 30076, "loss": 0.2527, "learning_rate": 2.0872380932240832e-05, "epoch": 1.4562380516997755, "percentage": 72.82, "elapsed_time": "2 days, 1:30:01", "remaining_time": "18:28:48", "throughput": "0.00", "total_tokens": 0}
|
| 222 |
+
{"current_steps": 22000, "total_steps": 30076, "loss": 0.2739, "learning_rate": 2.0402677365681112e-05, "epoch": 1.4628875405203225, "percentage": 73.15, "elapsed_time": "2 days, 1:43:01", "remaining_time": "18:15:02", "throughput": "0.00", "total_tokens": 0}
|
| 223 |
+
{"current_steps": 22100, "total_steps": 30076, "loss": 0.2823, "learning_rate": 1.99369606956191e-05, "epoch": 1.4695370293408694, "percentage": 73.48, "elapsed_time": "2 days, 1:56:06", "remaining_time": "18:01:18", "throughput": "0.00", "total_tokens": 0}
|
| 224 |
+
{"current_steps": 22200, "total_steps": 30076, "loss": 0.2753, "learning_rate": 1.9475293656248182e-05, "epoch": 1.4761865181614162, "percentage": 73.81, "elapsed_time": "2 days, 2:09:10", "remaining_time": "17:47:34", "throughput": "0.00", "total_tokens": 0}
|
| 225 |
+
{"current_steps": 22300, "total_steps": 30076, "loss": 0.2806, "learning_rate": 1.9017738436257655e-05, "epoch": 1.4828360069819633, "percentage": 74.15, "elapsed_time": "2 days, 2:21:56", "remaining_time": "17:33:45", "throughput": "0.00", "total_tokens": 0}
|
| 226 |
+
{"current_steps": 22400, "total_steps": 30076, "loss": 0.2749, "learning_rate": 1.8564356670455767e-05, "epoch": 1.48948549580251, "percentage": 74.48, "elapsed_time": "2 days, 2:35:21", "remaining_time": "17:20:09", "throughput": "0.00", "total_tokens": 0}
|
| 227 |
+
{"current_steps": 22500, "total_steps": 30076, "loss": 0.265, "learning_rate": 1.8115209431467074e-05, "epoch": 1.4961349846230572, "percentage": 74.81, "elapsed_time": "2 days, 2:48:56", "remaining_time": "17:06:36", "throughput": "0.00", "total_tokens": 0}
|
| 228 |
+
{"current_steps": 22600, "total_steps": 30076, "loss": 0.264, "learning_rate": 1.767035722150582e-05, "epoch": 1.502784473443604, "percentage": 75.14, "elapsed_time": "2 days, 3:01:56", "remaining_time": "16:52:52", "throughput": "0.00", "total_tokens": 0}
|
| 229 |
+
{"current_steps": 22700, "total_steps": 30076, "loss": 0.2587, "learning_rate": 1.7229859964225868e-05, "epoch": 1.509433962264151, "percentage": 75.48, "elapsed_time": "2 days, 3:14:44", "remaining_time": "16:39:05", "throughput": "0.00", "total_tokens": 0}
|
| 230 |
+
{"current_steps": 22800, "total_steps": 30076, "loss": 0.2638, "learning_rate": 1.679377699664884e-05, "epoch": 1.5160834510846979, "percentage": 75.81, "elapsed_time": "2 days, 3:27:44", "remaining_time": "16:25:22", "throughput": "0.00", "total_tokens": 0}
|
| 231 |
+
{"current_steps": 22900, "total_steps": 30076, "loss": 0.2718, "learning_rate": 1.6362167061171063e-05, "epoch": 1.5227329399052447, "percentage": 76.14, "elapsed_time": "2 days, 3:41:09", "remaining_time": "16:11:47", "throughput": "0.00", "total_tokens": 0}
|
| 232 |
+
{"current_steps": 23000, "total_steps": 30076, "loss": 0.2553, "learning_rate": 1.5935088297650674e-05, "epoch": 1.5293824287257918, "percentage": 76.47, "elapsed_time": "2 days, 3:54:45", "remaining_time": "15:58:15", "throughput": "0.00", "total_tokens": 0}
|
| 233 |
+
{"current_steps": 23100, "total_steps": 30076, "loss": 0.2569, "learning_rate": 1.551259823557602e-05, "epoch": 1.5360319175463386, "percentage": 76.81, "elapsed_time": "2 days, 4:07:18", "remaining_time": "15:44:25", "throughput": "0.00", "total_tokens": 0}
|
| 234 |
+
{"current_steps": 23200, "total_steps": 30076, "loss": 0.2474, "learning_rate": 1.509475378631603e-05, "epoch": 1.5426814063668854, "percentage": 77.14, "elapsed_time": "2 days, 4:20:30", "remaining_time": "15:30:46", "throughput": "0.00", "total_tokens": 0}
|
| 235 |
+
{"current_steps": 23300, "total_steps": 30076, "loss": 0.2618, "learning_rate": 1.468161123545413e-05, "epoch": 1.5493308951874325, "percentage": 77.47, "elapsed_time": "2 days, 4:33:17", "remaining_time": "15:17:01", "throughput": "0.00", "total_tokens": 0}
|
| 236 |
+
{"current_steps": 23400, "total_steps": 30076, "loss": 0.2412, "learning_rate": 1.4273226235206178e-05, "epoch": 1.5559803840079793, "percentage": 77.8, "elapsed_time": "2 days, 4:46:06", "remaining_time": "15:03:17", "throughput": "0.00", "total_tokens": 0}
|
| 237 |
+
{"current_steps": 23500, "total_steps": 30076, "loss": 0.234, "learning_rate": 1.3869653796923993e-05, "epoch": 1.5626298728285262, "percentage": 78.14, "elapsed_time": "2 days, 4:59:25", "remaining_time": "14:49:41", "throughput": "0.00", "total_tokens": 0}
|
| 238 |
+
{"current_steps": 23600, "total_steps": 30076, "loss": 0.2529, "learning_rate": 1.3470948283684925e-05, "epoch": 1.5692793616490732, "percentage": 78.47, "elapsed_time": "2 days, 5:11:44", "remaining_time": "14:35:50", "throughput": "0.00", "total_tokens": 0}
|
| 239 |
+
{"current_steps": 23700, "total_steps": 30076, "loss": 0.2444, "learning_rate": 1.307716340296904e-05, "epoch": 1.5759288504696203, "percentage": 78.8, "elapsed_time": "2 days, 5:25:02", "remaining_time": "14:22:15", "throughput": "0.00", "total_tokens": 0}
|
| 240 |
+
{"current_steps": 23800, "total_steps": 30076, "loss": 0.2533, "learning_rate": 1.268835219942433e-05, "epoch": 1.5825783392901671, "percentage": 79.13, "elapsed_time": "2 days, 5:38:24", "remaining_time": "14:08:41", "throughput": "0.00", "total_tokens": 0}
|
| 241 |
+
{"current_steps": 23900, "total_steps": 30076, "loss": 0.2467, "learning_rate": 1.23045670477215e-05, "epoch": 1.589227828110714, "percentage": 79.47, "elapsed_time": "2 days, 5:51:30", "remaining_time": "13:55:03", "throughput": "0.00", "total_tokens": 0}
|
| 242 |
+
{"current_steps": 24000, "total_steps": 30076, "loss": 0.2349, "learning_rate": 1.1925859645498722e-05, "epoch": 1.595877316931261, "percentage": 79.8, "elapsed_time": "2 days, 6:04:12", "remaining_time": "13:41:19", "throughput": "0.00", "total_tokens": 0}
|
| 243 |
+
{"current_steps": 24100, "total_steps": 30076, "loss": 0.2518, "learning_rate": 1.1552281006397819e-05, "epoch": 1.6025268057518078, "percentage": 80.13, "elapsed_time": "2 days, 6:16:49", "remaining_time": "13:27:35", "throughput": "0.00", "total_tokens": 0}
|
| 244 |
+
{"current_steps": 24200, "total_steps": 30076, "loss": 0.2376, "learning_rate": 1.1183881453192479e-05, "epoch": 1.6091762945723547, "percentage": 80.46, "elapsed_time": "2 days, 6:29:13", "remaining_time": "13:13:48", "throughput": "0.00", "total_tokens": 0}
|
| 245 |
+
{"current_steps": 24300, "total_steps": 30076, "loss": 0.2277, "learning_rate": 1.082071061100945e-05, "epoch": 1.6158257833929017, "percentage": 80.8, "elapsed_time": "2 days, 6:42:53", "remaining_time": "13:00:19", "throughput": "0.00", "total_tokens": 0}
|
| 246 |
+
{"current_steps": 24400, "total_steps": 30076, "loss": 0.2313, "learning_rate": 1.0462817400643959e-05, "epoch": 1.6224752722134486, "percentage": 81.13, "elapsed_time": "2 days, 6:55:18", "remaining_time": "12:46:33", "throughput": "0.00", "total_tokens": 0}
|
| 247 |
+
{"current_steps": 24500, "total_steps": 30076, "loss": 0.2364, "learning_rate": 1.0110250031969709e-05, "epoch": 1.6291247610339954, "percentage": 81.46, "elapsed_time": "2 days, 7:07:27", "remaining_time": "12:32:45", "throughput": "0.00", "total_tokens": 0}
|
| 248 |
+
{"current_steps": 24600, "total_steps": 30076, "loss": 0.2401, "learning_rate": 9.763055997444897e-06, "epoch": 1.6357742498545425, "percentage": 81.79, "elapsed_time": "2 days, 7:20:52", "remaining_time": "12:19:13", "throughput": "0.00", "total_tokens": 0}
|
| 249 |
+
{"current_steps": 24700, "total_steps": 30076, "loss": 0.2344, "learning_rate": 9.421282065714676e-06, "epoch": 1.6424237386750895, "percentage": 82.13, "elapsed_time": "2 days, 7:33:55", "remaining_time": "12:05:38", "throughput": "0.00", "total_tokens": 0}
|
| 250 |
+
{"current_steps": 24800, "total_steps": 30076, "loss": 0.2186, "learning_rate": 9.08497427531128e-06, "epoch": 1.6490732274956361, "percentage": 82.46, "elapsed_time": "2 days, 7:47:00", "remaining_time": "11:52:03", "throughput": "0.00", "total_tokens": 0}
|
| 251 |
+
{"current_steps": 24900, "total_steps": 30076, "loss": 0.2307, "learning_rate": 8.754177928452328e-06, "epoch": 1.6557227163161832, "percentage": 82.79, "elapsed_time": "2 days, 7:59:30", "remaining_time": "11:38:20", "throughput": "0.00", "total_tokens": 0}
|
| 252 |
+
{"current_steps": 25000, "total_steps": 30076, "loss": 0.2401, "learning_rate": 8.428937584938496e-06, "epoch": 1.6623722051367302, "percentage": 83.12, "elapsed_time": "2 days, 8:12:36", "remaining_time": "11:24:46", "throughput": "0.00", "total_tokens": 0}
|
| 253 |
+
{"current_steps": 25100, "total_steps": 30076, "loss": 0.2285, "learning_rate": 8.109297056151067e-06, "epoch": 1.669021693957277, "percentage": 83.46, "elapsed_time": "2 days, 8:25:48", "remaining_time": "11:11:13", "throughput": "0.00", "total_tokens": 0}
|
| 254 |
+
{"current_steps": 25200, "total_steps": 30076, "loss": 0.238, "learning_rate": 7.79529939915029e-06, "epoch": 1.675671182777824, "percentage": 83.79, "elapsed_time": "2 days, 8:38:49", "remaining_time": "10:57:38", "throughput": "0.00", "total_tokens": 0}
|
| 255 |
+
{"current_steps": 25300, "total_steps": 30076, "loss": 0.2245, "learning_rate": 7.486986910875499e-06, "epoch": 1.682320671598371, "percentage": 84.12, "elapsed_time": "2 days, 8:51:35", "remaining_time": "10:44:01", "throughput": "0.00", "total_tokens": 0}
|
| 256 |
+
{"current_steps": 25400, "total_steps": 30076, "loss": 0.2344, "learning_rate": 7.184401122447398e-06, "epoch": 1.6889701604189178, "percentage": 84.45, "elapsed_time": "2 days, 9:03:38", "remaining_time": "10:30:16", "throughput": "0.00", "total_tokens": 0}
|
| 257 |
+
{"current_steps": 25500, "total_steps": 30076, "loss": 0.2393, "learning_rate": 6.887582793573727e-06, "epoch": 1.6956196492394646, "percentage": 84.79, "elapsed_time": "2 days, 9:16:51", "remaining_time": "10:16:44", "throughput": "0.00", "total_tokens": 0}
|
| 258 |
+
{"current_steps": 25600, "total_steps": 30076, "loss": 0.2238, "learning_rate": 6.596571907058707e-06, "epoch": 1.7022691380600117, "percentage": 85.12, "elapsed_time": "2 days, 9:29:54", "remaining_time": "10:03:11", "throughput": "0.00", "total_tokens": 0}
|
| 259 |
+
{"current_steps": 25700, "total_steps": 30076, "loss": 0.2326, "learning_rate": 6.31140766341713e-06, "epoch": 1.7089186268805585, "percentage": 85.45, "elapsed_time": "2 days, 9:43:01", "remaining_time": "9:49:39", "throughput": "0.00", "total_tokens": 0}
|
| 260 |
+
{"current_steps": 25800, "total_steps": 30076, "loss": 0.2196, "learning_rate": 6.032128475593924e-06, "epoch": 1.7155681157011053, "percentage": 85.78, "elapsed_time": "2 days, 9:55:34", "remaining_time": "9:36:01", "throughput": "0.00", "total_tokens": 0}
|
| 261 |
+
{"current_steps": 25900, "total_steps": 30076, "loss": 0.2065, "learning_rate": 5.758771963789722e-06, "epoch": 1.7222176045216524, "percentage": 86.12, "elapsed_time": "2 days, 10:08:48", "remaining_time": "9:22:31", "throughput": "0.00", "total_tokens": 0}
|
| 262 |
+
{"current_steps": 26000, "total_steps": 30076, "loss": 0.2001, "learning_rate": 5.4913749503932575e-06, "epoch": 1.7288670933421995, "percentage": 86.45, "elapsed_time": "2 days, 10:21:43", "remaining_time": "9:08:57", "throughput": "0.00", "total_tokens": 0}
|
| 263 |
+
{"current_steps": 26100, "total_steps": 30076, "loss": 0.2183, "learning_rate": 5.229973455021231e-06, "epoch": 1.7355165821627463, "percentage": 86.78, "elapsed_time": "2 days, 10:35:09", "remaining_time": "8:55:29", "throughput": "0.00", "total_tokens": 0}
|
| 264 |
+
{"current_steps": 26200, "total_steps": 30076, "loss": 0.2106, "learning_rate": 4.974602689666252e-06, "epoch": 1.7421660709832931, "percentage": 87.11, "elapsed_time": "2 days, 10:47:50", "remaining_time": "8:41:54", "throughput": "0.00", "total_tokens": 0}
|
| 265 |
+
{"current_steps": 26300, "total_steps": 30076, "loss": 0.2397, "learning_rate": 4.725297053953692e-06, "epoch": 1.7488155598038402, "percentage": 87.45, "elapsed_time": "2 days, 11:00:14", "remaining_time": "8:28:17", "throughput": "0.00", "total_tokens": 0}
|
| 266 |
+
{"current_steps": 26400, "total_steps": 30076, "loss": 0.2209, "learning_rate": 4.48209013050781e-06, "epoch": 1.755465048624387, "percentage": 87.78, "elapsed_time": "2 days, 11:12:47", "remaining_time": "8:14:41", "throughput": "0.00", "total_tokens": 0}
|
| 267 |
+
{"current_steps": 26500, "total_steps": 30076, "loss": 0.211, "learning_rate": 4.245014680428117e-06, "epoch": 1.7621145374449338, "percentage": 88.11, "elapsed_time": "2 days, 11:25:17", "remaining_time": "8:01:06", "throughput": "0.00", "total_tokens": 0}
|
| 268 |
+
{"current_steps": 26600, "total_steps": 30076, "loss": 0.2134, "learning_rate": 4.014102638876205e-06, "epoch": 1.768764026265481, "percentage": 88.44, "elapsed_time": "2 days, 11:38:32", "remaining_time": "7:47:37", "throughput": "0.00", "total_tokens": 0}
|
| 269 |
+
{"current_steps": 26700, "total_steps": 30076, "loss": 0.2095, "learning_rate": 3.789385110774013e-06, "epoch": 1.7754135150860277, "percentage": 88.78, "elapsed_time": "2 days, 11:51:56", "remaining_time": "7:34:10", "throughput": "0.00", "total_tokens": 0}
|
| 270 |
+
{"current_steps": 26800, "total_steps": 30076, "loss": 0.238, "learning_rate": 3.5708923666137927e-06, "epoch": 1.7820630039065746, "percentage": 89.11, "elapsed_time": "2 days, 12:04:45", "remaining_time": "7:20:38", "throughput": "0.00", "total_tokens": 0}
|
| 271 |
+
{"current_steps": 26900, "total_steps": 30076, "loss": 0.2087, "learning_rate": 3.358653838380571e-06, "epoch": 1.7887124927271216, "percentage": 89.44, "elapsed_time": "2 days, 12:18:12", "remaining_time": "7:07:11", "throughput": "0.00", "total_tokens": 0}
|
| 272 |
+
{"current_steps": 27000, "total_steps": 30076, "loss": 0.2164, "learning_rate": 3.1526981155875156e-06, "epoch": 1.7953619815476687, "percentage": 89.77, "elapsed_time": "2 days, 12:31:17", "remaining_time": "6:53:41", "throughput": "0.00", "total_tokens": 0}
|
| 273 |
+
{"current_steps": 27100, "total_steps": 30076, "loss": 0.2143, "learning_rate": 2.9530529414247608e-06, "epoch": 1.8020114703682153, "percentage": 90.11, "elapsed_time": "2 days, 12:44:24", "remaining_time": "6:40:12", "throughput": "0.00", "total_tokens": 0}
|
| 274 |
+
{"current_steps": 27200, "total_steps": 30076, "loss": 0.2037, "learning_rate": 2.7597452090223354e-06, "epoch": 1.8086609591887624, "percentage": 90.44, "elapsed_time": "2 days, 12:57:28", "remaining_time": "6:26:43", "throughput": "0.00", "total_tokens": 0}
|
| 275 |
+
{"current_steps": 27300, "total_steps": 30076, "loss": 0.2156, "learning_rate": 2.572800957827476e-06, "epoch": 1.8153104480093094, "percentage": 90.77, "elapsed_time": "2 days, 13:10:52", "remaining_time": "6:13:16", "throughput": "0.00", "total_tokens": 0}
|
| 276 |
+
{"current_steps": 27400, "total_steps": 30076, "loss": 0.2182, "learning_rate": 2.3922453700970295e-06, "epoch": 1.8219599368298562, "percentage": 91.1, "elapsed_time": "2 days, 13:23:40", "remaining_time": "5:59:45", "throughput": "0.00", "total_tokens": 0}
|
| 277 |
+
{"current_steps": 27500, "total_steps": 30076, "loss": 0.2238, "learning_rate": 2.2181027675052534e-06, "epoch": 1.828609425650403, "percentage": 91.44, "elapsed_time": "2 days, 13:36:09", "remaining_time": "5:46:13", "throughput": "0.00", "total_tokens": 0}
|
| 278 |
+
{"current_steps": 27600, "total_steps": 30076, "loss": 0.2179, "learning_rate": 2.0503966078676217e-06, "epoch": 1.8352589144709501, "percentage": 91.77, "elapsed_time": "2 days, 13:49:10", "remaining_time": "5:32:44", "throughput": "0.00", "total_tokens": 0}
|
| 279 |
+
{"current_steps": 27700, "total_steps": 30076, "loss": 0.2149, "learning_rate": 1.8891494819808841e-06, "epoch": 1.841908403291497, "percentage": 92.1, "elapsed_time": "2 days, 14:02:15", "remaining_time": "5:19:16", "throughput": "0.00", "total_tokens": 0}
|
| 280 |
+
{"current_steps": 27800, "total_steps": 30076, "loss": 0.2071, "learning_rate": 1.7343831105800511e-06, "epoch": 1.8485578921120438, "percentage": 92.43, "elapsed_time": "2 days, 14:15:10", "remaining_time": "5:05:48", "throughput": "0.00", "total_tokens": 0}
|
| 281 |
+
{"current_steps": 27900, "total_steps": 30076, "loss": 0.2132, "learning_rate": 1.5861183414124403e-06, "epoch": 1.8552073809325909, "percentage": 92.76, "elapsed_time": "2 days, 14:28:43", "remaining_time": "4:52:22", "throughput": "0.00", "total_tokens": 0}
|
| 282 |
+
{"current_steps": 28000, "total_steps": 30076, "loss": 0.2097, "learning_rate": 1.4443751464294664e-06, "epoch": 1.8618568697531377, "percentage": 93.1, "elapsed_time": "2 days, 14:41:27", "remaining_time": "4:38:53", "throughput": "0.00", "total_tokens": 0}
|
| 283 |
+
{"current_steps": 28100, "total_steps": 30076, "loss": 0.1923, "learning_rate": 1.3091726190962329e-06, "epoch": 1.8685063585736845, "percentage": 93.43, "elapsed_time": "2 days, 14:53:59", "remaining_time": "4:25:23", "throughput": "0.00", "total_tokens": 0}
|
| 284 |
+
{"current_steps": 28200, "total_steps": 30076, "loss": 0.189, "learning_rate": 1.18052897181965e-06, "epoch": 1.8751558473942316, "percentage": 93.76, "elapsed_time": "2 days, 15:06:38", "remaining_time": "4:11:54", "throughput": "0.00", "total_tokens": 0}
|
| 285 |
+
{"current_steps": 28300, "total_steps": 30076, "loss": 0.2089, "learning_rate": 1.0584615334950643e-06, "epoch": 1.8818053362147786, "percentage": 94.09, "elapsed_time": "2 days, 15:19:41", "remaining_time": "3:58:27", "throughput": "0.00", "total_tokens": 0}
|
| 286 |
+
{"current_steps": 28400, "total_steps": 30076, "loss": 0.2045, "learning_rate": 9.429867471720255e-07, "epoch": 1.8884548250353252, "percentage": 94.43, "elapsed_time": "2 days, 15:32:22", "remaining_time": "3:44:59", "throughput": "0.00", "total_tokens": 0}
|
| 287 |
+
{"current_steps": 28500, "total_steps": 30076, "loss": 0.2235, "learning_rate": 8.341201678392974e-07, "epoch": 1.8951043138558723, "percentage": 94.76, "elapsed_time": "2 days, 15:45:03", "remaining_time": "3:31:31", "throughput": "0.00", "total_tokens": 0}
|
| 288 |
+
{"current_steps": 28600, "total_steps": 30076, "loss": 0.2037, "learning_rate": 7.318764603295447e-07, "epoch": 1.9017538026764194, "percentage": 95.09, "elapsed_time": "2 days, 15:58:04", "remaining_time": "3:18:04", "throughput": "0.00", "total_tokens": 0}
|
| 289 |
+
{"current_steps": 28700, "total_steps": 30076, "loss": 0.2108, "learning_rate": 6.362693973439193e-07, "epoch": 1.9084032914969662, "percentage": 95.42, "elapsed_time": "2 days, 16:10:48", "remaining_time": "3:04:37", "throughput": "0.00", "total_tokens": 0}
|
| 290 |
+
{"current_steps": 28800, "total_steps": 30076, "loss": 0.1984, "learning_rate": 5.47311857596794e-07, "epoch": 1.915052780317513, "percentage": 95.76, "elapsed_time": "2 days, 16:23:39", "remaining_time": "2:51:10", "throughput": "0.00", "total_tokens": 0}
|
| 291 |
+
{"current_steps": 28900, "total_steps": 30076, "loss": 0.2124, "learning_rate": 4.6501582408096657e-07, "epoch": 1.92170226913806, "percentage": 96.09, "elapsed_time": "2 days, 16:36:50", "remaining_time": "2:37:45", "throughput": "0.00", "total_tokens": 0}
|
| 292 |
+
{"current_steps": 29000, "total_steps": 30076, "loss": 0.1897, "learning_rate": 3.893923824534629e-07, "epoch": 1.928351757958607, "percentage": 96.42, "elapsed_time": "2 days, 16:49:04", "remaining_time": "2:24:17", "throughput": "0.00", "total_tokens": 0}
|
| 293 |
+
{"current_steps": 29100, "total_steps": 30076, "loss": 0.2267, "learning_rate": 3.204517195422696e-07, "epoch": 1.9350012467791537, "percentage": 96.75, "elapsed_time": "2 days, 17:01:44", "remaining_time": "2:10:51", "throughput": "0.00", "total_tokens": 0}
|
| 294 |
+
{"current_steps": 29200, "total_steps": 30076, "loss": 0.2104, "learning_rate": 2.5820312197411543e-07, "epoch": 1.9416507355997008, "percentage": 97.09, "elapsed_time": "2 days, 17:14:20", "remaining_time": "1:57:25", "throughput": "0.00", "total_tokens": 0}
|
| 295 |
+
{"current_steps": 29300, "total_steps": 30076, "loss": 0.2072, "learning_rate": 2.0265497492352735e-07, "epoch": 1.9483002244202476, "percentage": 97.42, "elapsed_time": "2 days, 17:27:25", "remaining_time": "1:44:00", "throughput": "0.00", "total_tokens": 0}
|
| 296 |
+
{"current_steps": 29400, "total_steps": 30076, "loss": 0.2018, "learning_rate": 1.538147609832896e-07, "epoch": 1.9549497132407945, "percentage": 97.75, "elapsed_time": "2 days, 17:41:13", "remaining_time": "1:30:37", "throughput": "0.00", "total_tokens": 0}
|
| 297 |
+
{"current_steps": 29500, "total_steps": 30076, "loss": 0.1963, "learning_rate": 1.1168905915652228e-07, "epoch": 1.9615992020613415, "percentage": 98.08, "elapsed_time": "2 days, 17:53:49", "remaining_time": "1:17:12", "throughput": "0.00", "total_tokens": 0}
|
| 298 |
+
{"current_steps": 29600, "total_steps": 30076, "loss": 0.2062, "learning_rate": 7.628354397045123e-08, "epoch": 1.9682486908818886, "percentage": 98.42, "elapsed_time": "2 days, 18:06:56", "remaining_time": "1:03:47", "throughput": "0.00", "total_tokens": 0}
|
| 299 |
+
{"current_steps": 29700, "total_steps": 30076, "loss": 0.2007, "learning_rate": 4.760298471201963e-08, "epoch": 1.9748981797024354, "percentage": 98.75, "elapsed_time": "2 days, 18:19:31", "remaining_time": "0:50:22", "throughput": "0.00", "total_tokens": 0}
|
| 300 |
+
{"current_steps": 29800, "total_steps": 30076, "loss": 0.2119, "learning_rate": 2.565124478545733e-08, "epoch": 1.9815476685229823, "percentage": 99.08, "elapsed_time": "2 days, 18:32:28", "remaining_time": "0:36:58", "throughput": "0.00", "total_tokens": 0}
|
| 301 |
+
{"current_steps": 29900, "total_steps": 30076, "loss": 0.2192, "learning_rate": 1.043128119184167e-08, "epoch": 1.9881971573435293, "percentage": 99.41, "elapsed_time": "2 days, 18:45:14", "remaining_time": "0:23:34", "throughput": "0.00", "total_tokens": 0}
|
| 302 |
+
{"current_steps": 30000, "total_steps": 30076, "loss": 0.2238, "learning_rate": 1.945144130788279e-09, "epoch": 1.9948466461640761, "percentage": 99.75, "elapsed_time": "2 days, 18:58:41", "remaining_time": "0:10:10", "throughput": "0.00", "total_tokens": 0}
|
| 303 |
+
{"current_steps": 30000, "total_steps": 30076, "eval_loss": 0.23713438212871552, "epoch": 1.9948466461640761, "percentage": 99.75, "elapsed_time": "2 days, 20:10:41", "remaining_time": "0:10:21", "throughput": "0.00", "total_tokens": 0}
|
| 304 |
+
{"current_steps": 30076, "total_steps": 30076, "epoch": 1.9999002576676919, "percentage": 100.0, "elapsed_time": "2 days, 20:20:39", "remaining_time": "0:00:00", "throughput": "0.00", "total_tokens": 0}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2166 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.9999002576676919,
|
| 5 |
+
"eval_steps": 10000,
|
| 6 |
+
"global_step": 30076,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.006649488820546921,
|
| 13 |
+
"grad_norm": 0.2602408528327942,
|
| 14 |
+
"learning_rate": 3.324468085106383e-06,
|
| 15 |
+
"loss": 0.818,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.013298977641093841,
|
| 20 |
+
"grad_norm": 0.14098776876926422,
|
| 21 |
+
"learning_rate": 6.648936170212766e-06,
|
| 22 |
+
"loss": 0.5412,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.019948466461640762,
|
| 27 |
+
"grad_norm": 0.4028046429157257,
|
| 28 |
+
"learning_rate": 9.973404255319148e-06,
|
| 29 |
+
"loss": 0.5292,
|
| 30 |
+
"step": 300
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.026597955282187682,
|
| 34 |
+
"grad_norm": 0.15940147638320923,
|
| 35 |
+
"learning_rate": 1.3297872340425532e-05,
|
| 36 |
+
"loss": 0.5203,
|
| 37 |
+
"step": 400
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0332474441027346,
|
| 41 |
+
"grad_norm": 0.32640570402145386,
|
| 42 |
+
"learning_rate": 1.6622340425531915e-05,
|
| 43 |
+
"loss": 0.5078,
|
| 44 |
+
"step": 500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.039896932923281524,
|
| 48 |
+
"grad_norm": 0.24910107254981995,
|
| 49 |
+
"learning_rate": 1.9946808510638297e-05,
|
| 50 |
+
"loss": 0.4875,
|
| 51 |
+
"step": 600
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.04654642174382844,
|
| 55 |
+
"grad_norm": 0.43389075994491577,
|
| 56 |
+
"learning_rate": 2.3271276595744682e-05,
|
| 57 |
+
"loss": 0.4978,
|
| 58 |
+
"step": 700
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.053195910564375365,
|
| 62 |
+
"grad_norm": 0.2782938778400421,
|
| 63 |
+
"learning_rate": 2.6595744680851064e-05,
|
| 64 |
+
"loss": 0.4807,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.05984539938492228,
|
| 69 |
+
"grad_norm": 0.3213005065917969,
|
| 70 |
+
"learning_rate": 2.9920212765957452e-05,
|
| 71 |
+
"loss": 0.5007,
|
| 72 |
+
"step": 900
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.0664948882054692,
|
| 76 |
+
"grad_norm": 0.7565635442733765,
|
| 77 |
+
"learning_rate": 3.324468085106383e-05,
|
| 78 |
+
"loss": 0.4754,
|
| 79 |
+
"step": 1000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.07314437702601613,
|
| 83 |
+
"grad_norm": 0.5631841421127319,
|
| 84 |
+
"learning_rate": 3.6569148936170215e-05,
|
| 85 |
+
"loss": 0.4943,
|
| 86 |
+
"step": 1100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.07979386584656305,
|
| 90 |
+
"grad_norm": 0.40041548013687134,
|
| 91 |
+
"learning_rate": 3.9893617021276594e-05,
|
| 92 |
+
"loss": 0.4928,
|
| 93 |
+
"step": 1200
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.08644335466710996,
|
| 97 |
+
"grad_norm": 0.43981432914733887,
|
| 98 |
+
"learning_rate": 4.3218085106382986e-05,
|
| 99 |
+
"loss": 0.4829,
|
| 100 |
+
"step": 1300
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.09309284348765688,
|
| 104 |
+
"grad_norm": 0.49957290291786194,
|
| 105 |
+
"learning_rate": 4.6542553191489364e-05,
|
| 106 |
+
"loss": 0.4797,
|
| 107 |
+
"step": 1400
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.09974233230820381,
|
| 111 |
+
"grad_norm": 1.0348299741744995,
|
| 112 |
+
"learning_rate": 4.986702127659575e-05,
|
| 113 |
+
"loss": 0.5052,
|
| 114 |
+
"step": 1500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.10639182112875073,
|
| 118 |
+
"grad_norm": 0.7030391097068787,
|
| 119 |
+
"learning_rate": 5.319148936170213e-05,
|
| 120 |
+
"loss": 0.4693,
|
| 121 |
+
"step": 1600
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.11304130994929765,
|
| 125 |
+
"grad_norm": 0.7447443008422852,
|
| 126 |
+
"learning_rate": 5.651595744680851e-05,
|
| 127 |
+
"loss": 0.4767,
|
| 128 |
+
"step": 1700
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.11969079876984456,
|
| 132 |
+
"grad_norm": 0.5579438209533691,
|
| 133 |
+
"learning_rate": 5.9840425531914904e-05,
|
| 134 |
+
"loss": 0.4823,
|
| 135 |
+
"step": 1800
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.12634028759039148,
|
| 139 |
+
"grad_norm": 0.5548936724662781,
|
| 140 |
+
"learning_rate": 6.316489361702128e-05,
|
| 141 |
+
"loss": 0.4631,
|
| 142 |
+
"step": 1900
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.1329897764109384,
|
| 146 |
+
"grad_norm": 0.9034917950630188,
|
| 147 |
+
"learning_rate": 6.648936170212766e-05,
|
| 148 |
+
"loss": 0.5277,
|
| 149 |
+
"step": 2000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.13963926523148534,
|
| 153 |
+
"grad_norm": 0.7092494964599609,
|
| 154 |
+
"learning_rate": 6.981382978723405e-05,
|
| 155 |
+
"loss": 0.4813,
|
| 156 |
+
"step": 2100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.14628875405203226,
|
| 160 |
+
"grad_norm": 1.0150775909423828,
|
| 161 |
+
"learning_rate": 7.313829787234043e-05,
|
| 162 |
+
"loss": 0.5026,
|
| 163 |
+
"step": 2200
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.15293824287257918,
|
| 167 |
+
"grad_norm": 1.1101422309875488,
|
| 168 |
+
"learning_rate": 7.646276595744682e-05,
|
| 169 |
+
"loss": 0.5226,
|
| 170 |
+
"step": 2300
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.1595877316931261,
|
| 174 |
+
"grad_norm": 0.5893256664276123,
|
| 175 |
+
"learning_rate": 7.978723404255319e-05,
|
| 176 |
+
"loss": 0.5285,
|
| 177 |
+
"step": 2400
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.166237220513673,
|
| 181 |
+
"grad_norm": 0.9117974042892456,
|
| 182 |
+
"learning_rate": 8.311170212765957e-05,
|
| 183 |
+
"loss": 0.5183,
|
| 184 |
+
"step": 2500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.17288670933421993,
|
| 188 |
+
"grad_norm": 1.3572144508361816,
|
| 189 |
+
"learning_rate": 8.643617021276597e-05,
|
| 190 |
+
"loss": 0.5129,
|
| 191 |
+
"step": 2600
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.17953619815476685,
|
| 195 |
+
"grad_norm": 0.890178382396698,
|
| 196 |
+
"learning_rate": 8.976063829787234e-05,
|
| 197 |
+
"loss": 0.5081,
|
| 198 |
+
"step": 2700
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.18618568697531376,
|
| 202 |
+
"grad_norm": 2.725203275680542,
|
| 203 |
+
"learning_rate": 9.308510638297873e-05,
|
| 204 |
+
"loss": 0.5042,
|
| 205 |
+
"step": 2800
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.19283517579586068,
|
| 209 |
+
"grad_norm": 0.962457001209259,
|
| 210 |
+
"learning_rate": 9.640957446808511e-05,
|
| 211 |
+
"loss": 0.5357,
|
| 212 |
+
"step": 2900
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.19948466461640763,
|
| 216 |
+
"grad_norm": 2.529735803604126,
|
| 217 |
+
"learning_rate": 9.97340425531915e-05,
|
| 218 |
+
"loss": 0.5407,
|
| 219 |
+
"step": 3000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.20613415343695454,
|
| 223 |
+
"grad_norm": 1.643441081047058,
|
| 224 |
+
"learning_rate": 9.999714964504067e-05,
|
| 225 |
+
"loss": 0.5244,
|
| 226 |
+
"step": 3100
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.21278364225750146,
|
| 230 |
+
"grad_norm": 0.9840033054351807,
|
| 231 |
+
"learning_rate": 9.998758599534463e-05,
|
| 232 |
+
"loss": 0.5562,
|
| 233 |
+
"step": 3200
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.21943313107804838,
|
| 237 |
+
"grad_norm": 1.0612396001815796,
|
| 238 |
+
"learning_rate": 9.997128878619808e-05,
|
| 239 |
+
"loss": 0.5273,
|
| 240 |
+
"step": 3300
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.2260826198985953,
|
| 244 |
+
"grad_norm": 1.1562029123306274,
|
| 245 |
+
"learning_rate": 9.994826021291056e-05,
|
| 246 |
+
"loss": 0.5412,
|
| 247 |
+
"step": 3400
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.2327321087191422,
|
| 251 |
+
"grad_norm": 1.5558404922485352,
|
| 252 |
+
"learning_rate": 9.991850337753762e-05,
|
| 253 |
+
"loss": 0.5405,
|
| 254 |
+
"step": 3500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.23938159753968913,
|
| 258 |
+
"grad_norm": 2.0897717475891113,
|
| 259 |
+
"learning_rate": 9.988202228846291e-05,
|
| 260 |
+
"loss": 0.5049,
|
| 261 |
+
"step": 3600
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.24603108636023605,
|
| 265 |
+
"grad_norm": 1.031879186630249,
|
| 266 |
+
"learning_rate": 9.983882185985808e-05,
|
| 267 |
+
"loss": 0.5785,
|
| 268 |
+
"step": 3700
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.25268057518078296,
|
| 272 |
+
"grad_norm": 1.3407084941864014,
|
| 273 |
+
"learning_rate": 9.978890791102109e-05,
|
| 274 |
+
"loss": 0.5487,
|
| 275 |
+
"step": 3800
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.2593300640013299,
|
| 279 |
+
"grad_norm": 1.261853814125061,
|
| 280 |
+
"learning_rate": 9.973228716559209e-05,
|
| 281 |
+
"loss": 0.533,
|
| 282 |
+
"step": 3900
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.2659795528218768,
|
| 286 |
+
"grad_norm": 1.1546040773391724,
|
| 287 |
+
"learning_rate": 9.966896725064786e-05,
|
| 288 |
+
"loss": 0.5166,
|
| 289 |
+
"step": 4000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.27262904164242374,
|
| 293 |
+
"grad_norm": 1.1395092010498047,
|
| 294 |
+
"learning_rate": 9.959895669567435e-05,
|
| 295 |
+
"loss": 0.5366,
|
| 296 |
+
"step": 4100
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.2792785304629707,
|
| 300 |
+
"grad_norm": 1.5521488189697266,
|
| 301 |
+
"learning_rate": 9.952226493141765e-05,
|
| 302 |
+
"loss": 0.5339,
|
| 303 |
+
"step": 4200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.2859280192835176,
|
| 307 |
+
"grad_norm": 1.9154760837554932,
|
| 308 |
+
"learning_rate": 9.943890228861383e-05,
|
| 309 |
+
"loss": 0.4976,
|
| 310 |
+
"step": 4300
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.2925775081040645,
|
| 314 |
+
"grad_norm": 1.273290753364563,
|
| 315 |
+
"learning_rate": 9.934887999659707e-05,
|
| 316 |
+
"loss": 0.5331,
|
| 317 |
+
"step": 4400
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.2992269969246114,
|
| 321 |
+
"grad_norm": 1.6757972240447998,
|
| 322 |
+
"learning_rate": 9.925221018178728e-05,
|
| 323 |
+
"loss": 0.5212,
|
| 324 |
+
"step": 4500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.30587648574515836,
|
| 328 |
+
"grad_norm": 1.3594906330108643,
|
| 329 |
+
"learning_rate": 9.914890586605638e-05,
|
| 330 |
+
"loss": 0.4991,
|
| 331 |
+
"step": 4600
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.31252597456570524,
|
| 335 |
+
"grad_norm": 1.622048258781433,
|
| 336 |
+
"learning_rate": 9.903898096497441e-05,
|
| 337 |
+
"loss": 0.569,
|
| 338 |
+
"step": 4700
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.3191754633862522,
|
| 342 |
+
"grad_norm": 1.6205965280532837,
|
| 343 |
+
"learning_rate": 9.892245028593483e-05,
|
| 344 |
+
"loss": 0.5436,
|
| 345 |
+
"step": 4800
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.3258249522067991,
|
| 349 |
+
"grad_norm": 2.7058024406433105,
|
| 350 |
+
"learning_rate": 9.879932952616009e-05,
|
| 351 |
+
"loss": 0.5028,
|
| 352 |
+
"step": 4900
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.332474441027346,
|
| 356 |
+
"grad_norm": 1.8897464275360107,
|
| 357 |
+
"learning_rate": 9.8669635270587e-05,
|
| 358 |
+
"loss": 0.7608,
|
| 359 |
+
"step": 5000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.33912392984789297,
|
| 363 |
+
"grad_norm": 0.9735029935836792,
|
| 364 |
+
"learning_rate": 9.853338498963272e-05,
|
| 365 |
+
"loss": 0.523,
|
| 366 |
+
"step": 5100
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.34577341866843986,
|
| 370 |
+
"grad_norm": 0.9950180053710938,
|
| 371 |
+
"learning_rate": 9.839059703684139e-05,
|
| 372 |
+
"loss": 0.5426,
|
| 373 |
+
"step": 5200
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.3524229074889868,
|
| 377 |
+
"grad_norm": 0.7262745499610901,
|
| 378 |
+
"learning_rate": 9.82412906464119e-05,
|
| 379 |
+
"loss": 0.5387,
|
| 380 |
+
"step": 5300
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.3590723963095337,
|
| 384 |
+
"grad_norm": 1.133631706237793,
|
| 385 |
+
"learning_rate": 9.808548593060681e-05,
|
| 386 |
+
"loss": 0.5259,
|
| 387 |
+
"step": 5400
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.36572188513008064,
|
| 391 |
+
"grad_norm": 1.8873803615570068,
|
| 392 |
+
"learning_rate": 9.792320387704328e-05,
|
| 393 |
+
"loss": 0.5363,
|
| 394 |
+
"step": 5500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.3723713739506275,
|
| 398 |
+
"grad_norm": 2.6183230876922607,
|
| 399 |
+
"learning_rate": 9.775446634586584e-05,
|
| 400 |
+
"loss": 0.542,
|
| 401 |
+
"step": 5600
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.37902086277117447,
|
| 405 |
+
"grad_norm": 1.7073196172714233,
|
| 406 |
+
"learning_rate": 9.757929606680181e-05,
|
| 407 |
+
"loss": 0.4947,
|
| 408 |
+
"step": 5700
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.38567035159172136,
|
| 412 |
+
"grad_norm": 1.0039314031600952,
|
| 413 |
+
"learning_rate": 9.73977166360994e-05,
|
| 414 |
+
"loss": 0.5584,
|
| 415 |
+
"step": 5800
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.3923198404122683,
|
| 419 |
+
"grad_norm": 1.0683766603469849,
|
| 420 |
+
"learning_rate": 9.720975251334929e-05,
|
| 421 |
+
"loss": 0.5318,
|
| 422 |
+
"step": 5900
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.39896932923281525,
|
| 426 |
+
"grad_norm": 1.4205585718154907,
|
| 427 |
+
"learning_rate": 9.701542901818974e-05,
|
| 428 |
+
"loss": 0.5352,
|
| 429 |
+
"step": 6000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.40561881805336214,
|
| 433 |
+
"grad_norm": 2.596090316772461,
|
| 434 |
+
"learning_rate": 9.681477232689596e-05,
|
| 435 |
+
"loss": 0.5229,
|
| 436 |
+
"step": 6100
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.4122683068739091,
|
| 440 |
+
"grad_norm": 1.1745169162750244,
|
| 441 |
+
"learning_rate": 9.660780946885397e-05,
|
| 442 |
+
"loss": 0.5419,
|
| 443 |
+
"step": 6200
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.418917795694456,
|
| 447 |
+
"grad_norm": 2.7921059131622314,
|
| 448 |
+
"learning_rate": 9.639456832291974e-05,
|
| 449 |
+
"loss": 0.5428,
|
| 450 |
+
"step": 6300
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.4255672845150029,
|
| 454 |
+
"grad_norm": 3.064069986343384,
|
| 455 |
+
"learning_rate": 9.617507761366367e-05,
|
| 456 |
+
"loss": 0.5256,
|
| 457 |
+
"step": 6400
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.4322167733355498,
|
| 461 |
+
"grad_norm": 1.322499394416809,
|
| 462 |
+
"learning_rate": 9.594936690750129e-05,
|
| 463 |
+
"loss": 0.5633,
|
| 464 |
+
"step": 6500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.43886626215609675,
|
| 468 |
+
"grad_norm": 1.6360127925872803,
|
| 469 |
+
"learning_rate": 9.571746660871058e-05,
|
| 470 |
+
"loss": 0.5269,
|
| 471 |
+
"step": 6600
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.44551575097664364,
|
| 475 |
+
"grad_norm": 1.442179799079895,
|
| 476 |
+
"learning_rate": 9.547940795533627e-05,
|
| 477 |
+
"loss": 0.5105,
|
| 478 |
+
"step": 6700
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.4521652397971906,
|
| 482 |
+
"grad_norm": 1.1154464483261108,
|
| 483 |
+
"learning_rate": 9.523522301498202e-05,
|
| 484 |
+
"loss": 0.5301,
|
| 485 |
+
"step": 6800
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.45881472861773753,
|
| 489 |
+
"grad_norm": 1.2995010614395142,
|
| 490 |
+
"learning_rate": 9.498494468049072e-05,
|
| 491 |
+
"loss": 0.5173,
|
| 492 |
+
"step": 6900
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.4654642174382844,
|
| 496 |
+
"grad_norm": 1.0487501621246338,
|
| 497 |
+
"learning_rate": 9.472860666551369e-05,
|
| 498 |
+
"loss": 0.5284,
|
| 499 |
+
"step": 7000
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.47211370625883137,
|
| 503 |
+
"grad_norm": 1.8207134008407593,
|
| 504 |
+
"learning_rate": 9.446624349996929e-05,
|
| 505 |
+
"loss": 0.5317,
|
| 506 |
+
"step": 7100
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.47876319507937826,
|
| 510 |
+
"grad_norm": 1.7722837924957275,
|
| 511 |
+
"learning_rate": 9.419789052539157e-05,
|
| 512 |
+
"loss": 0.5195,
|
| 513 |
+
"step": 7200
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.4854126838999252,
|
| 517 |
+
"grad_norm": 2.749199390411377,
|
| 518 |
+
"learning_rate": 9.392358389016961e-05,
|
| 519 |
+
"loss": 0.5274,
|
| 520 |
+
"step": 7300
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.4920621727204721,
|
| 524 |
+
"grad_norm": 1.3200327157974243,
|
| 525 |
+
"learning_rate": 9.364336054467819e-05,
|
| 526 |
+
"loss": 0.523,
|
| 527 |
+
"step": 7400
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.49871166154101904,
|
| 531 |
+
"grad_norm": 2.9674599170684814,
|
| 532 |
+
"learning_rate": 9.335725823630035e-05,
|
| 533 |
+
"loss": 0.5294,
|
| 534 |
+
"step": 7500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.5053611503615659,
|
| 538 |
+
"grad_norm": 2.8731188774108887,
|
| 539 |
+
"learning_rate": 9.306531550434268e-05,
|
| 540 |
+
"loss": 0.5087,
|
| 541 |
+
"step": 7600
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.5120106391821129,
|
| 545 |
+
"grad_norm": 2.215608596801758,
|
| 546 |
+
"learning_rate": 9.276757167484389e-05,
|
| 547 |
+
"loss": 0.5219,
|
| 548 |
+
"step": 7700
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.5186601280026598,
|
| 552 |
+
"grad_norm": 1.5501729249954224,
|
| 553 |
+
"learning_rate": 9.246406685527739e-05,
|
| 554 |
+
"loss": 0.5582,
|
| 555 |
+
"step": 7800
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.5253096168232068,
|
| 559 |
+
"grad_norm": 1.8816163539886475,
|
| 560 |
+
"learning_rate": 9.21548419291487e-05,
|
| 561 |
+
"loss": 0.5394,
|
| 562 |
+
"step": 7900
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.5319591056437536,
|
| 566 |
+
"grad_norm": 1.338460087776184,
|
| 567 |
+
"learning_rate": 9.183993855048811e-05,
|
| 568 |
+
"loss": 0.5299,
|
| 569 |
+
"step": 8000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.5386085944643005,
|
| 573 |
+
"grad_norm": 1.7432674169540405,
|
| 574 |
+
"learning_rate": 9.151939913823988e-05,
|
| 575 |
+
"loss": 0.5307,
|
| 576 |
+
"step": 8100
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.5452580832848475,
|
| 580 |
+
"grad_norm": 3.2521355152130127,
|
| 581 |
+
"learning_rate": 9.119326687054802e-05,
|
| 582 |
+
"loss": 0.5004,
|
| 583 |
+
"step": 8200
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.5519075721053944,
|
| 587 |
+
"grad_norm": 0.9576961994171143,
|
| 588 |
+
"learning_rate": 9.086158567894013e-05,
|
| 589 |
+
"loss": 0.5453,
|
| 590 |
+
"step": 8300
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.5585570609259414,
|
| 594 |
+
"grad_norm": 1.3667104244232178,
|
| 595 |
+
"learning_rate": 9.052440024240956e-05,
|
| 596 |
+
"loss": 0.5003,
|
| 597 |
+
"step": 8400
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.5652065497464882,
|
| 601 |
+
"grad_norm": 1.9807591438293457,
|
| 602 |
+
"learning_rate": 9.018175598139696e-05,
|
| 603 |
+
"loss": 0.508,
|
| 604 |
+
"step": 8500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.5718560385670352,
|
| 608 |
+
"grad_norm": 2.281437397003174,
|
| 609 |
+
"learning_rate": 8.983369905167191e-05,
|
| 610 |
+
"loss": 0.4788,
|
| 611 |
+
"step": 8600
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.5785055273875821,
|
| 615 |
+
"grad_norm": 2.3274779319763184,
|
| 616 |
+
"learning_rate": 8.948027633811557e-05,
|
| 617 |
+
"loss": 0.4994,
|
| 618 |
+
"step": 8700
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.585155016208129,
|
| 622 |
+
"grad_norm": 3.7350873947143555,
|
| 623 |
+
"learning_rate": 8.912153544840507e-05,
|
| 624 |
+
"loss": 0.546,
|
| 625 |
+
"step": 8800
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.5918045050286759,
|
| 629 |
+
"grad_norm": 1.8497984409332275,
|
| 630 |
+
"learning_rate": 8.875752470660043e-05,
|
| 631 |
+
"loss": 0.4939,
|
| 632 |
+
"step": 8900
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.5984539938492228,
|
| 636 |
+
"grad_norm": 2.2370221614837646,
|
| 637 |
+
"learning_rate": 8.838829314663522e-05,
|
| 638 |
+
"loss": 0.5288,
|
| 639 |
+
"step": 9000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.6051034826697698,
|
| 643 |
+
"grad_norm": 1.4384081363677979,
|
| 644 |
+
"learning_rate": 8.80138905057114e-05,
|
| 645 |
+
"loss": 0.5176,
|
| 646 |
+
"step": 9100
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.6117529714903167,
|
| 650 |
+
"grad_norm": 3.588320255279541,
|
| 651 |
+
"learning_rate": 8.763436721759952e-05,
|
| 652 |
+
"loss": 0.4898,
|
| 653 |
+
"step": 9200
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.6184024603108637,
|
| 657 |
+
"grad_norm": 2.6353256702423096,
|
| 658 |
+
"learning_rate": 8.724977440584497e-05,
|
| 659 |
+
"loss": 0.5011,
|
| 660 |
+
"step": 9300
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.6250519491314105,
|
| 664 |
+
"grad_norm": 1.7550867795944214,
|
| 665 |
+
"learning_rate": 8.686016387688153e-05,
|
| 666 |
+
"loss": 0.5101,
|
| 667 |
+
"step": 9400
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.6317014379519574,
|
| 671 |
+
"grad_norm": 1.9632377624511719,
|
| 672 |
+
"learning_rate": 8.646558811305268e-05,
|
| 673 |
+
"loss": 0.5023,
|
| 674 |
+
"step": 9500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.6383509267725044,
|
| 678 |
+
"grad_norm": 2.103327989578247,
|
| 679 |
+
"learning_rate": 8.60661002655421e-05,
|
| 680 |
+
"loss": 0.5055,
|
| 681 |
+
"step": 9600
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.6450004155930513,
|
| 685 |
+
"grad_norm": 2.518740653991699,
|
| 686 |
+
"learning_rate": 8.566175414721384e-05,
|
| 687 |
+
"loss": 0.4997,
|
| 688 |
+
"step": 9700
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.6516499044135982,
|
| 692 |
+
"grad_norm": 2.6153292655944824,
|
| 693 |
+
"learning_rate": 8.525260422536358e-05,
|
| 694 |
+
"loss": 0.5104,
|
| 695 |
+
"step": 9800
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.6582993932341451,
|
| 699 |
+
"grad_norm": 1.0710101127624512,
|
| 700 |
+
"learning_rate": 8.483870561438161e-05,
|
| 701 |
+
"loss": 0.5281,
|
| 702 |
+
"step": 9900
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.664948882054692,
|
| 706 |
+
"grad_norm": 3.5133893489837646,
|
| 707 |
+
"learning_rate": 8.442011406832859e-05,
|
| 708 |
+
"loss": 0.5147,
|
| 709 |
+
"step": 10000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.664948882054692,
|
| 713 |
+
"eval_loss": 0.4313640594482422,
|
| 714 |
+
"eval_runtime": 4340.2869,
|
| 715 |
+
"eval_samples_per_second": 1.54,
|
| 716 |
+
"eval_steps_per_second": 1.54,
|
| 717 |
+
"step": 10000
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.671598370875239,
|
| 721 |
+
"grad_norm": 1.416164755821228,
|
| 722 |
+
"learning_rate": 8.399688597342535e-05,
|
| 723 |
+
"loss": 0.5229,
|
| 724 |
+
"step": 10100
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.6782478596957859,
|
| 728 |
+
"grad_norm": 4.202746391296387,
|
| 729 |
+
"learning_rate": 8.356907834045726e-05,
|
| 730 |
+
"loss": 0.5008,
|
| 731 |
+
"step": 10200
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.6848973485163328,
|
| 735 |
+
"grad_norm": 1.6467583179473877,
|
| 736 |
+
"learning_rate": 8.313674879709475e-05,
|
| 737 |
+
"loss": 0.5084,
|
| 738 |
+
"step": 10300
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.6915468373368797,
|
| 742 |
+
"grad_norm": 0.9791653752326965,
|
| 743 |
+
"learning_rate": 8.269995558013049e-05,
|
| 744 |
+
"loss": 0.5049,
|
| 745 |
+
"step": 10400
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.6981963261574267,
|
| 749 |
+
"grad_norm": 1.2445042133331299,
|
| 750 |
+
"learning_rate": 8.225875752763468e-05,
|
| 751 |
+
"loss": 0.4711,
|
| 752 |
+
"step": 10500
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.7048458149779736,
|
| 756 |
+
"grad_norm": 3.4758522510528564,
|
| 757 |
+
"learning_rate": 8.181321407102929e-05,
|
| 758 |
+
"loss": 0.4835,
|
| 759 |
+
"step": 10600
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.7114953037985204,
|
| 763 |
+
"grad_norm": 1.5787672996520996,
|
| 764 |
+
"learning_rate": 8.136338522708233e-05,
|
| 765 |
+
"loss": 0.4849,
|
| 766 |
+
"step": 10700
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.7181447926190674,
|
| 770 |
+
"grad_norm": 2.1171529293060303,
|
| 771 |
+
"learning_rate": 8.090933158982338e-05,
|
| 772 |
+
"loss": 0.4881,
|
| 773 |
+
"step": 10800
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.7247942814396143,
|
| 777 |
+
"grad_norm": 3.9836864471435547,
|
| 778 |
+
"learning_rate": 8.045111432238121e-05,
|
| 779 |
+
"loss": 0.4699,
|
| 780 |
+
"step": 10900
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.7314437702601613,
|
| 784 |
+
"grad_norm": 3.3204433917999268,
|
| 785 |
+
"learning_rate": 7.998879514874491e-05,
|
| 786 |
+
"loss": 0.5022,
|
| 787 |
+
"step": 11000
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.7380932590807082,
|
| 791 |
+
"grad_norm": 1.5986028909683228,
|
| 792 |
+
"learning_rate": 7.952243634544929e-05,
|
| 793 |
+
"loss": 0.4933,
|
| 794 |
+
"step": 11100
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.744742747901255,
|
| 798 |
+
"grad_norm": 3.2928895950317383,
|
| 799 |
+
"learning_rate": 7.905210073318605e-05,
|
| 800 |
+
"loss": 0.4677,
|
| 801 |
+
"step": 11200
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.751392236721802,
|
| 805 |
+
"grad_norm": 2.922384738922119,
|
| 806 |
+
"learning_rate": 7.857785166834144e-05,
|
| 807 |
+
"loss": 0.4757,
|
| 808 |
+
"step": 11300
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.7580417255423489,
|
| 812 |
+
"grad_norm": 1.8758389949798584,
|
| 813 |
+
"learning_rate": 7.809975303446195e-05,
|
| 814 |
+
"loss": 0.4938,
|
| 815 |
+
"step": 11400
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.7646912143628959,
|
| 819 |
+
"grad_norm": 2.5382237434387207,
|
| 820 |
+
"learning_rate": 7.761786923364878e-05,
|
| 821 |
+
"loss": 0.4733,
|
| 822 |
+
"step": 11500
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.7713407031834427,
|
| 826 |
+
"grad_norm": 3.0387401580810547,
|
| 827 |
+
"learning_rate": 7.713226517788275e-05,
|
| 828 |
+
"loss": 0.4655,
|
| 829 |
+
"step": 11600
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.7779901920039897,
|
| 833 |
+
"grad_norm": 3.491082191467285,
|
| 834 |
+
"learning_rate": 7.664300628028017e-05,
|
| 835 |
+
"loss": 0.4933,
|
| 836 |
+
"step": 11700
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.7846396808245366,
|
| 840 |
+
"grad_norm": 2.891706943511963,
|
| 841 |
+
"learning_rate": 7.615015844628157e-05,
|
| 842 |
+
"loss": 0.4699,
|
| 843 |
+
"step": 11800
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.7912891696450836,
|
| 847 |
+
"grad_norm": 1.315558671951294,
|
| 848 |
+
"learning_rate": 7.565378806477377e-05,
|
| 849 |
+
"loss": 0.4514,
|
| 850 |
+
"step": 11900
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.7979386584656305,
|
| 854 |
+
"grad_norm": 5.145135402679443,
|
| 855 |
+
"learning_rate": 7.515396199914708e-05,
|
| 856 |
+
"loss": 0.4532,
|
| 857 |
+
"step": 12000
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.8045881472861773,
|
| 861 |
+
"grad_norm": 2.2307186126708984,
|
| 862 |
+
"learning_rate": 7.46507475782885e-05,
|
| 863 |
+
"loss": 0.4749,
|
| 864 |
+
"step": 12100
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.8112376361067243,
|
| 868 |
+
"grad_norm": 1.4436883926391602,
|
| 869 |
+
"learning_rate": 7.414421258751212e-05,
|
| 870 |
+
"loss": 0.4977,
|
| 871 |
+
"step": 12200
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.8178871249272712,
|
| 875 |
+
"grad_norm": 2.0418126583099365,
|
| 876 |
+
"learning_rate": 7.363442525942826e-05,
|
| 877 |
+
"loss": 0.4544,
|
| 878 |
+
"step": 12300
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.8245366137478182,
|
| 882 |
+
"grad_norm": 1.2745349407196045,
|
| 883 |
+
"learning_rate": 7.31214542647521e-05,
|
| 884 |
+
"loss": 0.5053,
|
| 885 |
+
"step": 12400
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.831186102568365,
|
| 889 |
+
"grad_norm": 3.4680609703063965,
|
| 890 |
+
"learning_rate": 7.260536870305347e-05,
|
| 891 |
+
"loss": 0.4648,
|
| 892 |
+
"step": 12500
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.837835591388912,
|
| 896 |
+
"grad_norm": 0.97030109167099,
|
| 897 |
+
"learning_rate": 7.208623809344879e-05,
|
| 898 |
+
"loss": 0.4313,
|
| 899 |
+
"step": 12600
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.8444850802094589,
|
| 903 |
+
"grad_norm": 3.7414844036102295,
|
| 904 |
+
"learning_rate": 7.156413236523656e-05,
|
| 905 |
+
"loss": 0.4816,
|
| 906 |
+
"step": 12700
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.8511345690300058,
|
| 910 |
+
"grad_norm": 1.8031582832336426,
|
| 911 |
+
"learning_rate": 7.103912184847757e-05,
|
| 912 |
+
"loss": 0.4534,
|
| 913 |
+
"step": 12800
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.8577840578505528,
|
| 917 |
+
"grad_norm": 1.6116315126419067,
|
| 918 |
+
"learning_rate": 7.051127726452102e-05,
|
| 919 |
+
"loss": 0.4304,
|
| 920 |
+
"step": 12900
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.8644335466710996,
|
| 924 |
+
"grad_norm": 2.230163335800171,
|
| 925 |
+
"learning_rate": 6.998066971647817e-05,
|
| 926 |
+
"loss": 0.4511,
|
| 927 |
+
"step": 13000
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.8710830354916466,
|
| 931 |
+
"grad_norm": 1.854151964187622,
|
| 932 |
+
"learning_rate": 6.944737067964429e-05,
|
| 933 |
+
"loss": 0.4412,
|
| 934 |
+
"step": 13100
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.8777325243121935,
|
| 938 |
+
"grad_norm": 2.757798194885254,
|
| 939 |
+
"learning_rate": 6.891145199187065e-05,
|
| 940 |
+
"loss": 0.4455,
|
| 941 |
+
"step": 13200
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.8843820131327405,
|
| 945 |
+
"grad_norm": 2.430884838104248,
|
| 946 |
+
"learning_rate": 6.837298584388771e-05,
|
| 947 |
+
"loss": 0.4598,
|
| 948 |
+
"step": 13300
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.8910315019532873,
|
| 952 |
+
"grad_norm": 1.6644638776779175,
|
| 953 |
+
"learning_rate": 6.783204476958058e-05,
|
| 954 |
+
"loss": 0.4561,
|
| 955 |
+
"step": 13400
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.8976809907738342,
|
| 959 |
+
"grad_norm": 2.7078986167907715,
|
| 960 |
+
"learning_rate": 6.728870163621836e-05,
|
| 961 |
+
"loss": 0.4818,
|
| 962 |
+
"step": 13500
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.9043304795943812,
|
| 966 |
+
"grad_norm": 1.8221591711044312,
|
| 967 |
+
"learning_rate": 6.674302963463876e-05,
|
| 968 |
+
"loss": 0.4322,
|
| 969 |
+
"step": 13600
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.9109799684149281,
|
| 973 |
+
"grad_norm": 1.0504103899002075,
|
| 974 |
+
"learning_rate": 6.61951022693887e-05,
|
| 975 |
+
"loss": 0.4601,
|
| 976 |
+
"step": 13700
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.9176294572354751,
|
| 980 |
+
"grad_norm": 1.788719892501831,
|
| 981 |
+
"learning_rate": 6.564499334882312e-05,
|
| 982 |
+
"loss": 0.4787,
|
| 983 |
+
"step": 13800
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.9242789460560219,
|
| 987 |
+
"grad_norm": 1.5331158638000488,
|
| 988 |
+
"learning_rate": 6.509277697516255e-05,
|
| 989 |
+
"loss": 0.4766,
|
| 990 |
+
"step": 13900
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.9309284348765688,
|
| 994 |
+
"grad_norm": 2.4608702659606934,
|
| 995 |
+
"learning_rate": 6.453852753451119e-05,
|
| 996 |
+
"loss": 0.4614,
|
| 997 |
+
"step": 14000
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.9375779236971158,
|
| 1001 |
+
"grad_norm": 1.432626485824585,
|
| 1002 |
+
"learning_rate": 6.39823196868368e-05,
|
| 1003 |
+
"loss": 0.48,
|
| 1004 |
+
"step": 14100
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.9442274125176627,
|
| 1008 |
+
"grad_norm": 3.4951376914978027,
|
| 1009 |
+
"learning_rate": 6.342422835591368e-05,
|
| 1010 |
+
"loss": 0.4717,
|
| 1011 |
+
"step": 14200
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.9508769013382097,
|
| 1015 |
+
"grad_norm": 1.7661138772964478,
|
| 1016 |
+
"learning_rate": 6.286432871923e-05,
|
| 1017 |
+
"loss": 0.4609,
|
| 1018 |
+
"step": 14300
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.9575263901587565,
|
| 1022 |
+
"grad_norm": 2.387974500656128,
|
| 1023 |
+
"learning_rate": 6.230269619786111e-05,
|
| 1024 |
+
"loss": 0.4233,
|
| 1025 |
+
"step": 14400
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.9641758789793035,
|
| 1029 |
+
"grad_norm": 1.5833674669265747,
|
| 1030 |
+
"learning_rate": 6.173940644630996e-05,
|
| 1031 |
+
"loss": 0.4368,
|
| 1032 |
+
"step": 14500
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.9708253677998504,
|
| 1036 |
+
"grad_norm": 2.6246190071105957,
|
| 1037 |
+
"learning_rate": 6.117453534231606e-05,
|
| 1038 |
+
"loss": 0.4205,
|
| 1039 |
+
"step": 14600
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.9774748566203973,
|
| 1043 |
+
"grad_norm": 1.6440967321395874,
|
| 1044 |
+
"learning_rate": 6.060815897663447e-05,
|
| 1045 |
+
"loss": 0.4247,
|
| 1046 |
+
"step": 14700
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.9841243454409442,
|
| 1050 |
+
"grad_norm": 2.0472023487091064,
|
| 1051 |
+
"learning_rate": 6.004035364278593e-05,
|
| 1052 |
+
"loss": 0.4338,
|
| 1053 |
+
"step": 14800
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.9907738342614911,
|
| 1057 |
+
"grad_norm": 3.3487770557403564,
|
| 1058 |
+
"learning_rate": 5.9471195826779834e-05,
|
| 1059 |
+
"loss": 0.4114,
|
| 1060 |
+
"step": 14900
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.9974233230820381,
|
| 1064 |
+
"grad_norm": 2.1973326206207275,
|
| 1065 |
+
"learning_rate": 5.8900762196811175e-05,
|
| 1066 |
+
"loss": 0.4209,
|
| 1067 |
+
"step": 15000
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.004072811902585,
|
| 1071 |
+
"grad_norm": 2.8700783252716064,
|
| 1072 |
+
"learning_rate": 5.832912959293304e-05,
|
| 1073 |
+
"loss": 0.3562,
|
| 1074 |
+
"step": 15100
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.0107223007231319,
|
| 1078 |
+
"grad_norm": 0.8708151578903198,
|
| 1079 |
+
"learning_rate": 5.775637501670579e-05,
|
| 1080 |
+
"loss": 0.3388,
|
| 1081 |
+
"step": 15200
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.017371789543679,
|
| 1085 |
+
"grad_norm": 0.9684904217720032,
|
| 1086 |
+
"learning_rate": 5.718257562082471e-05,
|
| 1087 |
+
"loss": 0.3572,
|
| 1088 |
+
"step": 15300
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.0240212783642257,
|
| 1092 |
+
"grad_norm": 1.5389324426651,
|
| 1093 |
+
"learning_rate": 5.660780869872711e-05,
|
| 1094 |
+
"loss": 0.3377,
|
| 1095 |
+
"step": 15400
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.0306707671847726,
|
| 1099 |
+
"grad_norm": 1.646532416343689,
|
| 1100 |
+
"learning_rate": 5.6032151674180575e-05,
|
| 1101 |
+
"loss": 0.3561,
|
| 1102 |
+
"step": 15500
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.0373202560053196,
|
| 1106 |
+
"grad_norm": 2.062817096710205,
|
| 1107 |
+
"learning_rate": 5.5455682090853624e-05,
|
| 1108 |
+
"loss": 0.3698,
|
| 1109 |
+
"step": 15600
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.0439697448258665,
|
| 1113 |
+
"grad_norm": 2.0568647384643555,
|
| 1114 |
+
"learning_rate": 5.4878477601870194e-05,
|
| 1115 |
+
"loss": 0.3223,
|
| 1116 |
+
"step": 15700
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.0506192336464135,
|
| 1120 |
+
"grad_norm": 2.101365566253662,
|
| 1121 |
+
"learning_rate": 5.430061595934941e-05,
|
| 1122 |
+
"loss": 0.3483,
|
| 1123 |
+
"step": 15800
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.0572687224669604,
|
| 1127 |
+
"grad_norm": 1.6456376314163208,
|
| 1128 |
+
"learning_rate": 5.372217500393205e-05,
|
| 1129 |
+
"loss": 0.3642,
|
| 1130 |
+
"step": 15900
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.0639182112875072,
|
| 1134 |
+
"grad_norm": 2.0329535007476807,
|
| 1135 |
+
"learning_rate": 5.314323265429501e-05,
|
| 1136 |
+
"loss": 0.3234,
|
| 1137 |
+
"step": 16000
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.0705677001080542,
|
| 1141 |
+
"grad_norm": 2.085597276687622,
|
| 1142 |
+
"learning_rate": 5.2563866896655275e-05,
|
| 1143 |
+
"loss": 0.3448,
|
| 1144 |
+
"step": 16100
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.077217188928601,
|
| 1148 |
+
"grad_norm": 2.2339513301849365,
|
| 1149 |
+
"learning_rate": 5.198415577426493e-05,
|
| 1150 |
+
"loss": 0.3621,
|
| 1151 |
+
"step": 16200
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.0838666777491481,
|
| 1155 |
+
"grad_norm": 2.7302517890930176,
|
| 1156 |
+
"learning_rate": 5.140417737689822e-05,
|
| 1157 |
+
"loss": 0.3532,
|
| 1158 |
+
"step": 16300
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.090516166569695,
|
| 1162 |
+
"grad_norm": 2.6171867847442627,
|
| 1163 |
+
"learning_rate": 5.0824009830332606e-05,
|
| 1164 |
+
"loss": 0.3424,
|
| 1165 |
+
"step": 16400
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.0971656553902418,
|
| 1169 |
+
"grad_norm": 1.7419236898422241,
|
| 1170 |
+
"learning_rate": 5.02437312858248e-05,
|
| 1171 |
+
"loss": 0.3465,
|
| 1172 |
+
"step": 16500
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.1038151442107889,
|
| 1176 |
+
"grad_norm": 1.149902582168579,
|
| 1177 |
+
"learning_rate": 4.966341990958347e-05,
|
| 1178 |
+
"loss": 0.3471,
|
| 1179 |
+
"step": 16600
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.1104646330313357,
|
| 1183 |
+
"grad_norm": 2.0462777614593506,
|
| 1184 |
+
"learning_rate": 4.908315387223985e-05,
|
| 1185 |
+
"loss": 0.3408,
|
| 1186 |
+
"step": 16700
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.1171141218518827,
|
| 1190 |
+
"grad_norm": 1.4826865196228027,
|
| 1191 |
+
"learning_rate": 4.850301133831786e-05,
|
| 1192 |
+
"loss": 0.3351,
|
| 1193 |
+
"step": 16800
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.1237636106724296,
|
| 1197 |
+
"grad_norm": 1.297524094581604,
|
| 1198 |
+
"learning_rate": 4.792307045570486e-05,
|
| 1199 |
+
"loss": 0.3531,
|
| 1200 |
+
"step": 16900
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.1304130994929764,
|
| 1204 |
+
"grad_norm": 0.9654291272163391,
|
| 1205 |
+
"learning_rate": 4.734340934512492e-05,
|
| 1206 |
+
"loss": 0.3171,
|
| 1207 |
+
"step": 17000
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.1370625883135235,
|
| 1211 |
+
"grad_norm": 1.5802520513534546,
|
| 1212 |
+
"learning_rate": 4.6764106089615454e-05,
|
| 1213 |
+
"loss": 0.3388,
|
| 1214 |
+
"step": 17100
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.1437120771340703,
|
| 1218 |
+
"grad_norm": 3.1234793663024902,
|
| 1219 |
+
"learning_rate": 4.618523872400921e-05,
|
| 1220 |
+
"loss": 0.3356,
|
| 1221 |
+
"step": 17200
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.1503615659546171,
|
| 1225 |
+
"grad_norm": 4.1535444259643555,
|
| 1226 |
+
"learning_rate": 4.56068852244225e-05,
|
| 1227 |
+
"loss": 0.3483,
|
| 1228 |
+
"step": 17300
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.1570110547751642,
|
| 1232 |
+
"grad_norm": 2.166757583618164,
|
| 1233 |
+
"learning_rate": 4.5029123497751514e-05,
|
| 1234 |
+
"loss": 0.3223,
|
| 1235 |
+
"step": 17400
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.163660543595711,
|
| 1239 |
+
"grad_norm": 1.9507828950881958,
|
| 1240 |
+
"learning_rate": 4.445203137117788e-05,
|
| 1241 |
+
"loss": 0.3181,
|
| 1242 |
+
"step": 17500
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.170310032416258,
|
| 1246 |
+
"grad_norm": 2.319566011428833,
|
| 1247 |
+
"learning_rate": 4.3875686581685e-05,
|
| 1248 |
+
"loss": 0.3416,
|
| 1249 |
+
"step": 17600
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.176959521236805,
|
| 1253 |
+
"grad_norm": 2.692155599594116,
|
| 1254 |
+
"learning_rate": 4.330016676558651e-05,
|
| 1255 |
+
"loss": 0.3269,
|
| 1256 |
+
"step": 17700
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.1836090100573518,
|
| 1260 |
+
"grad_norm": 1.3827095031738281,
|
| 1261 |
+
"learning_rate": 4.272554944806831e-05,
|
| 1262 |
+
"loss": 0.3166,
|
| 1263 |
+
"step": 17800
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.1902584988778988,
|
| 1267 |
+
"grad_norm": 1.681174635887146,
|
| 1268 |
+
"learning_rate": 4.2151912032745547e-05,
|
| 1269 |
+
"loss": 0.3354,
|
| 1270 |
+
"step": 17900
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.1969079876984456,
|
| 1274 |
+
"grad_norm": 1.9821159839630127,
|
| 1275 |
+
"learning_rate": 4.157933179123599e-05,
|
| 1276 |
+
"loss": 0.3066,
|
| 1277 |
+
"step": 18000
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.2035574765189927,
|
| 1281 |
+
"grad_norm": 1.9395629167556763,
|
| 1282 |
+
"learning_rate": 4.100788585275125e-05,
|
| 1283 |
+
"loss": 0.3427,
|
| 1284 |
+
"step": 18100
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.2102069653395395,
|
| 1288 |
+
"grad_norm": 1.5355767011642456,
|
| 1289 |
+
"learning_rate": 4.043765119370699e-05,
|
| 1290 |
+
"loss": 0.3074,
|
| 1291 |
+
"step": 18200
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.2168564541600864,
|
| 1295 |
+
"grad_norm": 2.809375047683716,
|
| 1296 |
+
"learning_rate": 3.9868704627354e-05,
|
| 1297 |
+
"loss": 0.3172,
|
| 1298 |
+
"step": 18300
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.2235059429806334,
|
| 1302 |
+
"grad_norm": 2.0123348236083984,
|
| 1303 |
+
"learning_rate": 3.930112279343094e-05,
|
| 1304 |
+
"loss": 0.3088,
|
| 1305 |
+
"step": 18400
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.2301554318011803,
|
| 1309 |
+
"grad_norm": 3.793424367904663,
|
| 1310 |
+
"learning_rate": 3.8734982147840756e-05,
|
| 1311 |
+
"loss": 0.3053,
|
| 1312 |
+
"step": 18500
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.236804920621727,
|
| 1316 |
+
"grad_norm": 3.0886518955230713,
|
| 1317 |
+
"learning_rate": 3.817035895235159e-05,
|
| 1318 |
+
"loss": 0.3178,
|
| 1319 |
+
"step": 18600
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.2434544094422741,
|
| 1323 |
+
"grad_norm": 3.0276455879211426,
|
| 1324 |
+
"learning_rate": 3.760732926432407e-05,
|
| 1325 |
+
"loss": 0.2876,
|
| 1326 |
+
"step": 18700
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.250103898262821,
|
| 1330 |
+
"grad_norm": 2.8517942428588867,
|
| 1331 |
+
"learning_rate": 3.704596892646593e-05,
|
| 1332 |
+
"loss": 0.3264,
|
| 1333 |
+
"step": 18800
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.256753387083368,
|
| 1337 |
+
"grad_norm": 3.9238364696502686,
|
| 1338 |
+
"learning_rate": 3.648635355661577e-05,
|
| 1339 |
+
"loss": 0.3151,
|
| 1340 |
+
"step": 18900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.2634028759039149,
|
| 1344 |
+
"grad_norm": 2.554077625274658,
|
| 1345 |
+
"learning_rate": 3.5928558537556895e-05,
|
| 1346 |
+
"loss": 0.3086,
|
| 1347 |
+
"step": 19000
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.270052364724462,
|
| 1351 |
+
"grad_norm": 1.959659457206726,
|
| 1352 |
+
"learning_rate": 3.537265900686286e-05,
|
| 1353 |
+
"loss": 0.2933,
|
| 1354 |
+
"step": 19100
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.2767018535450088,
|
| 1358 |
+
"grad_norm": 2.120922327041626,
|
| 1359 |
+
"learning_rate": 3.4818729846776254e-05,
|
| 1360 |
+
"loss": 0.2867,
|
| 1361 |
+
"step": 19200
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.2833513423655556,
|
| 1365 |
+
"grad_norm": 2.423426389694214,
|
| 1366 |
+
"learning_rate": 3.426684567412153e-05,
|
| 1367 |
+
"loss": 0.2956,
|
| 1368 |
+
"step": 19300
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.2900008311861026,
|
| 1372 |
+
"grad_norm": 2.462615489959717,
|
| 1373 |
+
"learning_rate": 3.371708083025392e-05,
|
| 1374 |
+
"loss": 0.2944,
|
| 1375 |
+
"step": 19400
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.2966503200066495,
|
| 1379 |
+
"grad_norm": 1.9235759973526,
|
| 1380 |
+
"learning_rate": 3.316950937104518e-05,
|
| 1381 |
+
"loss": 0.2868,
|
| 1382 |
+
"step": 19500
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.3032998088271963,
|
| 1386 |
+
"grad_norm": 3.4119069576263428,
|
| 1387 |
+
"learning_rate": 3.26242050569081e-05,
|
| 1388 |
+
"loss": 0.306,
|
| 1389 |
+
"step": 19600
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.3099492976477434,
|
| 1393 |
+
"grad_norm": 5.942444801330566,
|
| 1394 |
+
"learning_rate": 3.208124134286038e-05,
|
| 1395 |
+
"loss": 0.294,
|
| 1396 |
+
"step": 19700
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.3165987864682902,
|
| 1400 |
+
"grad_norm": 1.254133701324463,
|
| 1401 |
+
"learning_rate": 3.1540691368630185e-05,
|
| 1402 |
+
"loss": 0.3237,
|
| 1403 |
+
"step": 19800
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.323248275288837,
|
| 1407 |
+
"grad_norm": 3.514387369155884,
|
| 1408 |
+
"learning_rate": 3.100262794880363e-05,
|
| 1409 |
+
"loss": 0.3111,
|
| 1410 |
+
"step": 19900
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.329897764109384,
|
| 1414 |
+
"grad_norm": 7.018650054931641,
|
| 1415 |
+
"learning_rate": 3.0467123563016513e-05,
|
| 1416 |
+
"loss": 0.3106,
|
| 1417 |
+
"step": 20000
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.329897764109384,
|
| 1421 |
+
"eval_loss": 0.31190410256385803,
|
| 1422 |
+
"eval_runtime": 4327.0173,
|
| 1423 |
+
"eval_samples_per_second": 1.545,
|
| 1424 |
+
"eval_steps_per_second": 1.545,
|
| 1425 |
+
"step": 20000
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 1.336547252929931,
|
| 1429 |
+
"grad_norm": 0.9992861747741699,
|
| 1430 |
+
"learning_rate": 2.9934250346190818e-05,
|
| 1431 |
+
"loss": 0.2899,
|
| 1432 |
+
"step": 20100
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 1.343196741750478,
|
| 1436 |
+
"grad_norm": 2.0007855892181396,
|
| 1437 |
+
"learning_rate": 2.9404080078817924e-05,
|
| 1438 |
+
"loss": 0.284,
|
| 1439 |
+
"step": 20200
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 1.3498462305710248,
|
| 1443 |
+
"grad_norm": 1.0622339248657227,
|
| 1444 |
+
"learning_rate": 2.8876684177289404e-05,
|
| 1445 |
+
"loss": 0.2886,
|
| 1446 |
+
"step": 20300
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 1.3564957193915719,
|
| 1450 |
+
"grad_norm": 1.351544737815857,
|
| 1451 |
+
"learning_rate": 2.8352133684276853e-05,
|
| 1452 |
+
"loss": 0.2738,
|
| 1453 |
+
"step": 20400
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 1.3631452082121187,
|
| 1457 |
+
"grad_norm": 1.9642298221588135,
|
| 1458 |
+
"learning_rate": 2.7830499259162213e-05,
|
| 1459 |
+
"loss": 0.2981,
|
| 1460 |
+
"step": 20500
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 1.3697946970326655,
|
| 1464 |
+
"grad_norm": 1.769395351409912,
|
| 1465 |
+
"learning_rate": 2.7311851168519496e-05,
|
| 1466 |
+
"loss": 0.3159,
|
| 1467 |
+
"step": 20600
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 1.3764441858532126,
|
| 1471 |
+
"grad_norm": 3.744534492492676,
|
| 1472 |
+
"learning_rate": 2.6796259276649693e-05,
|
| 1473 |
+
"loss": 0.3019,
|
| 1474 |
+
"step": 20700
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 1.3830936746737594,
|
| 1478 |
+
"grad_norm": 0.7485169768333435,
|
| 1479 |
+
"learning_rate": 2.6283793036169603e-05,
|
| 1480 |
+
"loss": 0.3029,
|
| 1481 |
+
"step": 20800
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 1.3897431634943063,
|
| 1485 |
+
"grad_norm": 1.2193599939346313,
|
| 1486 |
+
"learning_rate": 2.5774521478656343e-05,
|
| 1487 |
+
"loss": 0.2799,
|
| 1488 |
+
"step": 20900
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 1.3963926523148533,
|
| 1492 |
+
"grad_norm": 3.9265201091766357,
|
| 1493 |
+
"learning_rate": 2.52685132053484e-05,
|
| 1494 |
+
"loss": 0.2783,
|
| 1495 |
+
"step": 21000
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 1.4030421411354002,
|
| 1499 |
+
"grad_norm": 1.0576444864273071,
|
| 1500 |
+
"learning_rate": 2.4765836377904787e-05,
|
| 1501 |
+
"loss": 0.3043,
|
| 1502 |
+
"step": 21100
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 1.4096916299559472,
|
| 1506 |
+
"grad_norm": 1.704397201538086,
|
| 1507 |
+
"learning_rate": 2.4266558709223293e-05,
|
| 1508 |
+
"loss": 0.2823,
|
| 1509 |
+
"step": 21200
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 1.416341118776494,
|
| 1513 |
+
"grad_norm": 5.58804178237915,
|
| 1514 |
+
"learning_rate": 2.377074745431931e-05,
|
| 1515 |
+
"loss": 0.2899,
|
| 1516 |
+
"step": 21300
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 1.422990607597041,
|
| 1520 |
+
"grad_norm": 2.1619229316711426,
|
| 1521 |
+
"learning_rate": 2.3278469401266178e-05,
|
| 1522 |
+
"loss": 0.259,
|
| 1523 |
+
"step": 21400
|
| 1524 |
+
},
|
| 1525 |
+
{
|
| 1526 |
+
"epoch": 1.429640096417588,
|
| 1527 |
+
"grad_norm": 3.6991875171661377,
|
| 1528 |
+
"learning_rate": 2.2789790862198628e-05,
|
| 1529 |
+
"loss": 0.2508,
|
| 1530 |
+
"step": 21500
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 1.4362895852381348,
|
| 1534 |
+
"grad_norm": 0.9216620922088623,
|
| 1535 |
+
"learning_rate": 2.2304777664380176e-05,
|
| 1536 |
+
"loss": 0.2701,
|
| 1537 |
+
"step": 21600
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 1.4429390740586818,
|
| 1541 |
+
"grad_norm": 1.0544836521148682,
|
| 1542 |
+
"learning_rate": 2.182349514133583e-05,
|
| 1543 |
+
"loss": 0.2741,
|
| 1544 |
+
"step": 21700
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 1.4495885628792287,
|
| 1548 |
+
"grad_norm": 1.3279638290405273,
|
| 1549 |
+
"learning_rate": 2.134600812405151e-05,
|
| 1550 |
+
"loss": 0.2608,
|
| 1551 |
+
"step": 21800
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 1.4562380516997755,
|
| 1555 |
+
"grad_norm": 2.69641375541687,
|
| 1556 |
+
"learning_rate": 2.0872380932240832e-05,
|
| 1557 |
+
"loss": 0.2527,
|
| 1558 |
+
"step": 21900
|
| 1559 |
+
},
|
| 1560 |
+
{
|
| 1561 |
+
"epoch": 1.4628875405203225,
|
| 1562 |
+
"grad_norm": 1.3837428092956543,
|
| 1563 |
+
"learning_rate": 2.0402677365681112e-05,
|
| 1564 |
+
"loss": 0.2739,
|
| 1565 |
+
"step": 22000
|
| 1566 |
+
},
|
| 1567 |
+
{
|
| 1568 |
+
"epoch": 1.4695370293408694,
|
| 1569 |
+
"grad_norm": 2.859133720397949,
|
| 1570 |
+
"learning_rate": 1.99369606956191e-05,
|
| 1571 |
+
"loss": 0.2823,
|
| 1572 |
+
"step": 22100
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 1.4761865181614162,
|
| 1576 |
+
"grad_norm": 2.0277671813964844,
|
| 1577 |
+
"learning_rate": 1.9475293656248182e-05,
|
| 1578 |
+
"loss": 0.2753,
|
| 1579 |
+
"step": 22200
|
| 1580 |
+
},
|
| 1581 |
+
{
|
| 1582 |
+
"epoch": 1.4828360069819633,
|
| 1583 |
+
"grad_norm": 1.3588882684707642,
|
| 1584 |
+
"learning_rate": 1.9017738436257655e-05,
|
| 1585 |
+
"loss": 0.2806,
|
| 1586 |
+
"step": 22300
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 1.48948549580251,
|
| 1590 |
+
"grad_norm": 0.9804733991622925,
|
| 1591 |
+
"learning_rate": 1.8564356670455767e-05,
|
| 1592 |
+
"loss": 0.2749,
|
| 1593 |
+
"step": 22400
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 1.4961349846230572,
|
| 1597 |
+
"grad_norm": 1.8412333726882935,
|
| 1598 |
+
"learning_rate": 1.8115209431467074e-05,
|
| 1599 |
+
"loss": 0.265,
|
| 1600 |
+
"step": 22500
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 1.502784473443604,
|
| 1604 |
+
"grad_norm": 0.9384289979934692,
|
| 1605 |
+
"learning_rate": 1.767035722150582e-05,
|
| 1606 |
+
"loss": 0.264,
|
| 1607 |
+
"step": 22600
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 1.509433962264151,
|
| 1611 |
+
"grad_norm": 1.8639055490493774,
|
| 1612 |
+
"learning_rate": 1.7229859964225868e-05,
|
| 1613 |
+
"loss": 0.2587,
|
| 1614 |
+
"step": 22700
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 1.5160834510846979,
|
| 1618 |
+
"grad_norm": 1.7392961978912354,
|
| 1619 |
+
"learning_rate": 1.679377699664884e-05,
|
| 1620 |
+
"loss": 0.2638,
|
| 1621 |
+
"step": 22800
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 1.5227329399052447,
|
| 1625 |
+
"grad_norm": 2.0196003913879395,
|
| 1626 |
+
"learning_rate": 1.6362167061171063e-05,
|
| 1627 |
+
"loss": 0.2718,
|
| 1628 |
+
"step": 22900
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 1.5293824287257918,
|
| 1632 |
+
"grad_norm": 1.6492582559585571,
|
| 1633 |
+
"learning_rate": 1.5935088297650674e-05,
|
| 1634 |
+
"loss": 0.2553,
|
| 1635 |
+
"step": 23000
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 1.5360319175463386,
|
| 1639 |
+
"grad_norm": 1.1431901454925537,
|
| 1640 |
+
"learning_rate": 1.551259823557602e-05,
|
| 1641 |
+
"loss": 0.2569,
|
| 1642 |
+
"step": 23100
|
| 1643 |
+
},
|
| 1644 |
+
{
|
| 1645 |
+
"epoch": 1.5426814063668854,
|
| 1646 |
+
"grad_norm": 1.3804975748062134,
|
| 1647 |
+
"learning_rate": 1.509475378631603e-05,
|
| 1648 |
+
"loss": 0.2474,
|
| 1649 |
+
"step": 23200
|
| 1650 |
+
},
|
| 1651 |
+
{
|
| 1652 |
+
"epoch": 1.5493308951874325,
|
| 1653 |
+
"grad_norm": 1.6143451929092407,
|
| 1654 |
+
"learning_rate": 1.468161123545413e-05,
|
| 1655 |
+
"loss": 0.2618,
|
| 1656 |
+
"step": 23300
|
| 1657 |
+
},
|
| 1658 |
+
{
|
| 1659 |
+
"epoch": 1.5559803840079793,
|
| 1660 |
+
"grad_norm": 2.3701083660125732,
|
| 1661 |
+
"learning_rate": 1.4273226235206178e-05,
|
| 1662 |
+
"loss": 0.2412,
|
| 1663 |
+
"step": 23400
|
| 1664 |
+
},
|
| 1665 |
+
{
|
| 1666 |
+
"epoch": 1.5626298728285262,
|
| 1667 |
+
"grad_norm": 0.5906326174736023,
|
| 1668 |
+
"learning_rate": 1.3869653796923993e-05,
|
| 1669 |
+
"loss": 0.234,
|
| 1670 |
+
"step": 23500
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 1.5692793616490732,
|
| 1674 |
+
"grad_norm": 0.9088567495346069,
|
| 1675 |
+
"learning_rate": 1.3470948283684925e-05,
|
| 1676 |
+
"loss": 0.2529,
|
| 1677 |
+
"step": 23600
|
| 1678 |
+
},
|
| 1679 |
+
{
|
| 1680 |
+
"epoch": 1.5759288504696203,
|
| 1681 |
+
"grad_norm": 2.244845151901245,
|
| 1682 |
+
"learning_rate": 1.307716340296904e-05,
|
| 1683 |
+
"loss": 0.2444,
|
| 1684 |
+
"step": 23700
|
| 1685 |
+
},
|
| 1686 |
+
{
|
| 1687 |
+
"epoch": 1.5825783392901671,
|
| 1688 |
+
"grad_norm": 4.36794376373291,
|
| 1689 |
+
"learning_rate": 1.268835219942433e-05,
|
| 1690 |
+
"loss": 0.2533,
|
| 1691 |
+
"step": 23800
|
| 1692 |
+
},
|
| 1693 |
+
{
|
| 1694 |
+
"epoch": 1.589227828110714,
|
| 1695 |
+
"grad_norm": 2.707139730453491,
|
| 1696 |
+
"learning_rate": 1.23045670477215e-05,
|
| 1697 |
+
"loss": 0.2467,
|
| 1698 |
+
"step": 23900
|
| 1699 |
+
},
|
| 1700 |
+
{
|
| 1701 |
+
"epoch": 1.595877316931261,
|
| 1702 |
+
"grad_norm": 1.5950795412063599,
|
| 1703 |
+
"learning_rate": 1.1925859645498722e-05,
|
| 1704 |
+
"loss": 0.2349,
|
| 1705 |
+
"step": 24000
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 1.6025268057518078,
|
| 1709 |
+
"grad_norm": 1.1506330966949463,
|
| 1710 |
+
"learning_rate": 1.1552281006397819e-05,
|
| 1711 |
+
"loss": 0.2518,
|
| 1712 |
+
"step": 24100
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 1.6091762945723547,
|
| 1716 |
+
"grad_norm": 2.356943130493164,
|
| 1717 |
+
"learning_rate": 1.1183881453192479e-05,
|
| 1718 |
+
"loss": 0.2376,
|
| 1719 |
+
"step": 24200
|
| 1720 |
+
},
|
| 1721 |
+
{
|
| 1722 |
+
"epoch": 1.6158257833929017,
|
| 1723 |
+
"grad_norm": 1.1734323501586914,
|
| 1724 |
+
"learning_rate": 1.082071061100945e-05,
|
| 1725 |
+
"loss": 0.2277,
|
| 1726 |
+
"step": 24300
|
| 1727 |
+
},
|
| 1728 |
+
{
|
| 1729 |
+
"epoch": 1.6224752722134486,
|
| 1730 |
+
"grad_norm": 3.4108047485351562,
|
| 1731 |
+
"learning_rate": 1.0462817400643959e-05,
|
| 1732 |
+
"loss": 0.2313,
|
| 1733 |
+
"step": 24400
|
| 1734 |
+
},
|
| 1735 |
+
{
|
| 1736 |
+
"epoch": 1.6291247610339954,
|
| 1737 |
+
"grad_norm": 2.890157699584961,
|
| 1738 |
+
"learning_rate": 1.0110250031969709e-05,
|
| 1739 |
+
"loss": 0.2364,
|
| 1740 |
+
"step": 24500
|
| 1741 |
+
},
|
| 1742 |
+
{
|
| 1743 |
+
"epoch": 1.6357742498545425,
|
| 1744 |
+
"grad_norm": 1.4353581666946411,
|
| 1745 |
+
"learning_rate": 9.763055997444897e-06,
|
| 1746 |
+
"loss": 0.2401,
|
| 1747 |
+
"step": 24600
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"epoch": 1.6424237386750895,
|
| 1751 |
+
"grad_norm": 0.6841472387313843,
|
| 1752 |
+
"learning_rate": 9.421282065714676e-06,
|
| 1753 |
+
"loss": 0.2344,
|
| 1754 |
+
"step": 24700
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 1.6490732274956361,
|
| 1758 |
+
"grad_norm": 2.0864779949188232,
|
| 1759 |
+
"learning_rate": 9.08497427531128e-06,
|
| 1760 |
+
"loss": 0.2186,
|
| 1761 |
+
"step": 24800
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 1.6557227163161832,
|
| 1765 |
+
"grad_norm": 1.5929943323135376,
|
| 1766 |
+
"learning_rate": 8.754177928452328e-06,
|
| 1767 |
+
"loss": 0.2307,
|
| 1768 |
+
"step": 24900
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"epoch": 1.6623722051367302,
|
| 1772 |
+
"grad_norm": 0.5336441993713379,
|
| 1773 |
+
"learning_rate": 8.428937584938496e-06,
|
| 1774 |
+
"loss": 0.2401,
|
| 1775 |
+
"step": 25000
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 1.669021693957277,
|
| 1779 |
+
"grad_norm": 2.829430103302002,
|
| 1780 |
+
"learning_rate": 8.109297056151067e-06,
|
| 1781 |
+
"loss": 0.2285,
|
| 1782 |
+
"step": 25100
|
| 1783 |
+
},
|
| 1784 |
+
{
|
| 1785 |
+
"epoch": 1.675671182777824,
|
| 1786 |
+
"grad_norm": 1.822135090827942,
|
| 1787 |
+
"learning_rate": 7.79529939915029e-06,
|
| 1788 |
+
"loss": 0.238,
|
| 1789 |
+
"step": 25200
|
| 1790 |
+
},
|
| 1791 |
+
{
|
| 1792 |
+
"epoch": 1.682320671598371,
|
| 1793 |
+
"grad_norm": 3.5365030765533447,
|
| 1794 |
+
"learning_rate": 7.486986910875499e-06,
|
| 1795 |
+
"loss": 0.2245,
|
| 1796 |
+
"step": 25300
|
| 1797 |
+
},
|
| 1798 |
+
{
|
| 1799 |
+
"epoch": 1.6889701604189178,
|
| 1800 |
+
"grad_norm": 3.5611343383789062,
|
| 1801 |
+
"learning_rate": 7.184401122447398e-06,
|
| 1802 |
+
"loss": 0.2344,
|
| 1803 |
+
"step": 25400
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 1.6956196492394646,
|
| 1807 |
+
"grad_norm": 3.247079372406006,
|
| 1808 |
+
"learning_rate": 6.887582793573727e-06,
|
| 1809 |
+
"loss": 0.2393,
|
| 1810 |
+
"step": 25500
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 1.7022691380600117,
|
| 1814 |
+
"grad_norm": 2.570157766342163,
|
| 1815 |
+
"learning_rate": 6.596571907058707e-06,
|
| 1816 |
+
"loss": 0.2238,
|
| 1817 |
+
"step": 25600
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 1.7089186268805585,
|
| 1821 |
+
"grad_norm": 1.488516092300415,
|
| 1822 |
+
"learning_rate": 6.31140766341713e-06,
|
| 1823 |
+
"loss": 0.2326,
|
| 1824 |
+
"step": 25700
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 1.7155681157011053,
|
| 1828 |
+
"grad_norm": 1.449952483177185,
|
| 1829 |
+
"learning_rate": 6.032128475593924e-06,
|
| 1830 |
+
"loss": 0.2196,
|
| 1831 |
+
"step": 25800
|
| 1832 |
+
},
|
| 1833 |
+
{
|
| 1834 |
+
"epoch": 1.7222176045216524,
|
| 1835 |
+
"grad_norm": 2.537747621536255,
|
| 1836 |
+
"learning_rate": 5.758771963789722e-06,
|
| 1837 |
+
"loss": 0.2065,
|
| 1838 |
+
"step": 25900
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 1.7288670933421995,
|
| 1842 |
+
"grad_norm": 1.783013105392456,
|
| 1843 |
+
"learning_rate": 5.4913749503932575e-06,
|
| 1844 |
+
"loss": 0.2001,
|
| 1845 |
+
"step": 26000
|
| 1846 |
+
},
|
| 1847 |
+
{
|
| 1848 |
+
"epoch": 1.7355165821627463,
|
| 1849 |
+
"grad_norm": 2.2559077739715576,
|
| 1850 |
+
"learning_rate": 5.229973455021231e-06,
|
| 1851 |
+
"loss": 0.2183,
|
| 1852 |
+
"step": 26100
|
| 1853 |
+
},
|
| 1854 |
+
{
|
| 1855 |
+
"epoch": 1.7421660709832931,
|
| 1856 |
+
"grad_norm": 2.048455238342285,
|
| 1857 |
+
"learning_rate": 4.974602689666252e-06,
|
| 1858 |
+
"loss": 0.2106,
|
| 1859 |
+
"step": 26200
|
| 1860 |
+
},
|
| 1861 |
+
{
|
| 1862 |
+
"epoch": 1.7488155598038402,
|
| 1863 |
+
"grad_norm": 1.1111422777175903,
|
| 1864 |
+
"learning_rate": 4.725297053953692e-06,
|
| 1865 |
+
"loss": 0.2397,
|
| 1866 |
+
"step": 26300
|
| 1867 |
+
},
|
| 1868 |
+
{
|
| 1869 |
+
"epoch": 1.755465048624387,
|
| 1870 |
+
"grad_norm": 1.549456238746643,
|
| 1871 |
+
"learning_rate": 4.48209013050781e-06,
|
| 1872 |
+
"loss": 0.2209,
|
| 1873 |
+
"step": 26400
|
| 1874 |
+
},
|
| 1875 |
+
{
|
| 1876 |
+
"epoch": 1.7621145374449338,
|
| 1877 |
+
"grad_norm": 2.2229185104370117,
|
| 1878 |
+
"learning_rate": 4.245014680428117e-06,
|
| 1879 |
+
"loss": 0.211,
|
| 1880 |
+
"step": 26500
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 1.768764026265481,
|
| 1884 |
+
"grad_norm": 3.065992593765259,
|
| 1885 |
+
"learning_rate": 4.014102638876205e-06,
|
| 1886 |
+
"loss": 0.2134,
|
| 1887 |
+
"step": 26600
|
| 1888 |
+
},
|
| 1889 |
+
{
|
| 1890 |
+
"epoch": 1.7754135150860277,
|
| 1891 |
+
"grad_norm": 1.1763139963150024,
|
| 1892 |
+
"learning_rate": 3.789385110774013e-06,
|
| 1893 |
+
"loss": 0.2095,
|
| 1894 |
+
"step": 26700
|
| 1895 |
+
},
|
| 1896 |
+
{
|
| 1897 |
+
"epoch": 1.7820630039065746,
|
| 1898 |
+
"grad_norm": 2.860395908355713,
|
| 1899 |
+
"learning_rate": 3.5708923666137927e-06,
|
| 1900 |
+
"loss": 0.238,
|
| 1901 |
+
"step": 26800
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 1.7887124927271216,
|
| 1905 |
+
"grad_norm": 2.1108574867248535,
|
| 1906 |
+
"learning_rate": 3.358653838380571e-06,
|
| 1907 |
+
"loss": 0.2087,
|
| 1908 |
+
"step": 26900
|
| 1909 |
+
},
|
| 1910 |
+
{
|
| 1911 |
+
"epoch": 1.7953619815476687,
|
| 1912 |
+
"grad_norm": 0.8311635255813599,
|
| 1913 |
+
"learning_rate": 3.1526981155875156e-06,
|
| 1914 |
+
"loss": 0.2164,
|
| 1915 |
+
"step": 27000
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 1.8020114703682153,
|
| 1919 |
+
"grad_norm": 0.8638597130775452,
|
| 1920 |
+
"learning_rate": 2.9530529414247608e-06,
|
| 1921 |
+
"loss": 0.2143,
|
| 1922 |
+
"step": 27100
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 1.8086609591887624,
|
| 1926 |
+
"grad_norm": 3.8020806312561035,
|
| 1927 |
+
"learning_rate": 2.7597452090223354e-06,
|
| 1928 |
+
"loss": 0.2037,
|
| 1929 |
+
"step": 27200
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 1.8153104480093094,
|
| 1933 |
+
"grad_norm": 4.387086868286133,
|
| 1934 |
+
"learning_rate": 2.572800957827476e-06,
|
| 1935 |
+
"loss": 0.2156,
|
| 1936 |
+
"step": 27300
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 1.8219599368298562,
|
| 1940 |
+
"grad_norm": 1.4160668849945068,
|
| 1941 |
+
"learning_rate": 2.3922453700970295e-06,
|
| 1942 |
+
"loss": 0.2182,
|
| 1943 |
+
"step": 27400
|
| 1944 |
+
},
|
| 1945 |
+
{
|
| 1946 |
+
"epoch": 1.828609425650403,
|
| 1947 |
+
"grad_norm": 1.636501669883728,
|
| 1948 |
+
"learning_rate": 2.2181027675052534e-06,
|
| 1949 |
+
"loss": 0.2238,
|
| 1950 |
+
"step": 27500
|
| 1951 |
+
},
|
| 1952 |
+
{
|
| 1953 |
+
"epoch": 1.8352589144709501,
|
| 1954 |
+
"grad_norm": 2.2405593395233154,
|
| 1955 |
+
"learning_rate": 2.0503966078676217e-06,
|
| 1956 |
+
"loss": 0.2179,
|
| 1957 |
+
"step": 27600
|
| 1958 |
+
},
|
| 1959 |
+
{
|
| 1960 |
+
"epoch": 1.841908403291497,
|
| 1961 |
+
"grad_norm": 1.6398348808288574,
|
| 1962 |
+
"learning_rate": 1.8891494819808841e-06,
|
| 1963 |
+
"loss": 0.2149,
|
| 1964 |
+
"step": 27700
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 1.8485578921120438,
|
| 1968 |
+
"grad_norm": 3.231600522994995,
|
| 1969 |
+
"learning_rate": 1.7343831105800511e-06,
|
| 1970 |
+
"loss": 0.2071,
|
| 1971 |
+
"step": 27800
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"epoch": 1.8552073809325909,
|
| 1975 |
+
"grad_norm": 2.70027232170105,
|
| 1976 |
+
"learning_rate": 1.5861183414124403e-06,
|
| 1977 |
+
"loss": 0.2132,
|
| 1978 |
+
"step": 27900
|
| 1979 |
+
},
|
| 1980 |
+
{
|
| 1981 |
+
"epoch": 1.8618568697531377,
|
| 1982 |
+
"grad_norm": 2.990867853164673,
|
| 1983 |
+
"learning_rate": 1.4443751464294664e-06,
|
| 1984 |
+
"loss": 0.2097,
|
| 1985 |
+
"step": 28000
|
| 1986 |
+
},
|
| 1987 |
+
{
|
| 1988 |
+
"epoch": 1.8685063585736845,
|
| 1989 |
+
"grad_norm": 1.75977623462677,
|
| 1990 |
+
"learning_rate": 1.3091726190962329e-06,
|
| 1991 |
+
"loss": 0.1923,
|
| 1992 |
+
"step": 28100
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 1.8751558473942316,
|
| 1996 |
+
"grad_norm": 0.49737560749053955,
|
| 1997 |
+
"learning_rate": 1.18052897181965e-06,
|
| 1998 |
+
"loss": 0.189,
|
| 1999 |
+
"step": 28200
|
| 2000 |
+
},
|
| 2001 |
+
{
|
| 2002 |
+
"epoch": 1.8818053362147786,
|
| 2003 |
+
"grad_norm": 1.1605116128921509,
|
| 2004 |
+
"learning_rate": 1.0584615334950643e-06,
|
| 2005 |
+
"loss": 0.2089,
|
| 2006 |
+
"step": 28300
|
| 2007 |
+
},
|
| 2008 |
+
{
|
| 2009 |
+
"epoch": 1.8884548250353252,
|
| 2010 |
+
"grad_norm": 2.284116268157959,
|
| 2011 |
+
"learning_rate": 9.429867471720255e-07,
|
| 2012 |
+
"loss": 0.2045,
|
| 2013 |
+
"step": 28400
|
| 2014 |
+
},
|
| 2015 |
+
{
|
| 2016 |
+
"epoch": 1.8951043138558723,
|
| 2017 |
+
"grad_norm": 2.2018344402313232,
|
| 2018 |
+
"learning_rate": 8.341201678392974e-07,
|
| 2019 |
+
"loss": 0.2235,
|
| 2020 |
+
"step": 28500
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 1.9017538026764194,
|
| 2024 |
+
"grad_norm": 1.3855736255645752,
|
| 2025 |
+
"learning_rate": 7.318764603295447e-07,
|
| 2026 |
+
"loss": 0.2037,
|
| 2027 |
+
"step": 28600
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"epoch": 1.9084032914969662,
|
| 2031 |
+
"grad_norm": 2.4615190029144287,
|
| 2032 |
+
"learning_rate": 6.362693973439193e-07,
|
| 2033 |
+
"loss": 0.2108,
|
| 2034 |
+
"step": 28700
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"epoch": 1.915052780317513,
|
| 2038 |
+
"grad_norm": 2.775005578994751,
|
| 2039 |
+
"learning_rate": 5.47311857596794e-07,
|
| 2040 |
+
"loss": 0.1984,
|
| 2041 |
+
"step": 28800
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 1.92170226913806,
|
| 2045 |
+
"grad_norm": 3.7345826625823975,
|
| 2046 |
+
"learning_rate": 4.6501582408096657e-07,
|
| 2047 |
+
"loss": 0.2124,
|
| 2048 |
+
"step": 28900
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 1.928351757958607,
|
| 2052 |
+
"grad_norm": 0.46711668372154236,
|
| 2053 |
+
"learning_rate": 3.893923824534629e-07,
|
| 2054 |
+
"loss": 0.1897,
|
| 2055 |
+
"step": 29000
|
| 2056 |
+
},
|
| 2057 |
+
{
|
| 2058 |
+
"epoch": 1.9350012467791537,
|
| 2059 |
+
"grad_norm": 2.135247230529785,
|
| 2060 |
+
"learning_rate": 3.204517195422696e-07,
|
| 2061 |
+
"loss": 0.2267,
|
| 2062 |
+
"step": 29100
|
| 2063 |
+
},
|
| 2064 |
+
{
|
| 2065 |
+
"epoch": 1.9416507355997008,
|
| 2066 |
+
"grad_norm": 0.9539075493812561,
|
| 2067 |
+
"learning_rate": 2.5820312197411543e-07,
|
| 2068 |
+
"loss": 0.2104,
|
| 2069 |
+
"step": 29200
|
| 2070 |
+
},
|
| 2071 |
+
{
|
| 2072 |
+
"epoch": 1.9483002244202476,
|
| 2073 |
+
"grad_norm": 3.765820026397705,
|
| 2074 |
+
"learning_rate": 2.0265497492352735e-07,
|
| 2075 |
+
"loss": 0.2072,
|
| 2076 |
+
"step": 29300
|
| 2077 |
+
},
|
| 2078 |
+
{
|
| 2079 |
+
"epoch": 1.9549497132407945,
|
| 2080 |
+
"grad_norm": 2.848856210708618,
|
| 2081 |
+
"learning_rate": 1.538147609832896e-07,
|
| 2082 |
+
"loss": 0.2018,
|
| 2083 |
+
"step": 29400
|
| 2084 |
+
},
|
| 2085 |
+
{
|
| 2086 |
+
"epoch": 1.9615992020613415,
|
| 2087 |
+
"grad_norm": 1.4174009561538696,
|
| 2088 |
+
"learning_rate": 1.1168905915652228e-07,
|
| 2089 |
+
"loss": 0.1963,
|
| 2090 |
+
"step": 29500
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 1.9682486908818886,
|
| 2094 |
+
"grad_norm": 3.067931652069092,
|
| 2095 |
+
"learning_rate": 7.628354397045123e-08,
|
| 2096 |
+
"loss": 0.2062,
|
| 2097 |
+
"step": 29600
|
| 2098 |
+
},
|
| 2099 |
+
{
|
| 2100 |
+
"epoch": 1.9748981797024354,
|
| 2101 |
+
"grad_norm": 2.1927988529205322,
|
| 2102 |
+
"learning_rate": 4.760298471201963e-08,
|
| 2103 |
+
"loss": 0.2007,
|
| 2104 |
+
"step": 29700
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"epoch": 1.9815476685229823,
|
| 2108 |
+
"grad_norm": 2.1186487674713135,
|
| 2109 |
+
"learning_rate": 2.565124478545733e-08,
|
| 2110 |
+
"loss": 0.2119,
|
| 2111 |
+
"step": 29800
|
| 2112 |
+
},
|
| 2113 |
+
{
|
| 2114 |
+
"epoch": 1.9881971573435293,
|
| 2115 |
+
"grad_norm": 1.58423912525177,
|
| 2116 |
+
"learning_rate": 1.043128119184167e-08,
|
| 2117 |
+
"loss": 0.2192,
|
| 2118 |
+
"step": 29900
|
| 2119 |
+
},
|
| 2120 |
+
{
|
| 2121 |
+
"epoch": 1.9948466461640761,
|
| 2122 |
+
"grad_norm": 1.9130926132202148,
|
| 2123 |
+
"learning_rate": 1.945144130788279e-09,
|
| 2124 |
+
"loss": 0.2238,
|
| 2125 |
+
"step": 30000
|
| 2126 |
+
},
|
| 2127 |
+
{
|
| 2128 |
+
"epoch": 1.9948466461640761,
|
| 2129 |
+
"eval_loss": 0.23713438212871552,
|
| 2130 |
+
"eval_runtime": 4319.7371,
|
| 2131 |
+
"eval_samples_per_second": 1.547,
|
| 2132 |
+
"eval_steps_per_second": 1.547,
|
| 2133 |
+
"step": 30000
|
| 2134 |
+
},
|
| 2135 |
+
{
|
| 2136 |
+
"epoch": 1.9999002576676919,
|
| 2137 |
+
"step": 30076,
|
| 2138 |
+
"total_flos": 1.2039888836550205e+19,
|
| 2139 |
+
"train_loss": 0.385877049085006,
|
| 2140 |
+
"train_runtime": 246042.9971,
|
| 2141 |
+
"train_samples_per_second": 0.489,
|
| 2142 |
+
"train_steps_per_second": 0.122
|
| 2143 |
+
}
|
| 2144 |
+
],
|
| 2145 |
+
"logging_steps": 100,
|
| 2146 |
+
"max_steps": 30076,
|
| 2147 |
+
"num_input_tokens_seen": 0,
|
| 2148 |
+
"num_train_epochs": 2,
|
| 2149 |
+
"save_steps": 10000,
|
| 2150 |
+
"stateful_callbacks": {
|
| 2151 |
+
"TrainerControl": {
|
| 2152 |
+
"args": {
|
| 2153 |
+
"should_epoch_stop": false,
|
| 2154 |
+
"should_evaluate": false,
|
| 2155 |
+
"should_log": false,
|
| 2156 |
+
"should_save": true,
|
| 2157 |
+
"should_training_stop": true
|
| 2158 |
+
},
|
| 2159 |
+
"attributes": {}
|
| 2160 |
+
}
|
| 2161 |
+
},
|
| 2162 |
+
"total_flos": 1.2039888836550205e+19,
|
| 2163 |
+
"train_batch_size": 1,
|
| 2164 |
+
"trial_name": null,
|
| 2165 |
+
"trial_params": null
|
| 2166 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5cfa842d25fa8351f352f02583aae45a21ced2a0a987be7b662d36ab5390e8af
|
| 3 |
+
size 5432
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|