Commit
•
25e3320
1
Parent(s):
5a3713e
willherbert27/bert-combined-textbook
Browse files- README.md +76 -0
- config.json +26 -0
- generation_config.json +5 -0
- model.safetensors +3 -0
- trainer_state.json +1569 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: google-bert/bert-base-cased
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
model-index:
|
| 7 |
+
- name: bert-finetuned-combo-textbook
|
| 8 |
+
results: []
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 12 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 13 |
+
|
| 14 |
+
# bert-finetuned-combo-textbook
|
| 15 |
+
|
| 16 |
+
This model is a fine-tuned version of [google-bert/bert-base-cased](https://huggingface.co/google-bert/bert-base-cased) on the None dataset.
|
| 17 |
+
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 1.5965
|
| 19 |
+
|
| 20 |
+
## Model description
|
| 21 |
+
|
| 22 |
+
More information needed
|
| 23 |
+
|
| 24 |
+
## Intended uses & limitations
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Training and evaluation data
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training procedure
|
| 33 |
+
|
| 34 |
+
### Training hyperparameters
|
| 35 |
+
|
| 36 |
+
The following hyperparameters were used during training:
|
| 37 |
+
- learning_rate: 1e-05
|
| 38 |
+
- train_batch_size: 8
|
| 39 |
+
- eval_batch_size: 8
|
| 40 |
+
- seed: 42
|
| 41 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 42 |
+
- lr_scheduler_type: linear
|
| 43 |
+
- num_epochs: 20
|
| 44 |
+
|
| 45 |
+
### Training results
|
| 46 |
+
|
| 47 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 48 |
+
|:-------------:|:-----:|:-----:|:---------------:|
|
| 49 |
+
| 2.2161 | 1.0 | 4936 | nan |
|
| 50 |
+
| 2.0531 | 2.0 | 9872 | 1.9590 |
|
| 51 |
+
| 1.978 | 3.0 | 14808 | nan |
|
| 52 |
+
| 1.9616 | 4.0 | 19744 | nan |
|
| 53 |
+
| 1.8559 | 5.0 | 24680 | 1.8338 |
|
| 54 |
+
| 1.7902 | 6.0 | 29616 | 1.7772 |
|
| 55 |
+
| 1.7603 | 7.0 | 34552 | 1.7686 |
|
| 56 |
+
| 1.769 | 8.0 | 39488 | 1.7368 |
|
| 57 |
+
| 1.6527 | 9.0 | 44424 | nan |
|
| 58 |
+
| 1.65 | 10.0 | 49360 | 1.6888 |
|
| 59 |
+
| 1.6281 | 11.0 | 54296 | nan |
|
| 60 |
+
| 1.591 | 12.0 | 59232 | 1.6576 |
|
| 61 |
+
| 1.5508 | 13.0 | 64168 | nan |
|
| 62 |
+
| 1.5929 | 14.0 | 69104 | 1.6151 |
|
| 63 |
+
| 1.5504 | 15.0 | 74040 | nan |
|
| 64 |
+
| 1.5492 | 16.0 | 78976 | 1.6250 |
|
| 65 |
+
| 1.5488 | 17.0 | 83912 | 1.6311 |
|
| 66 |
+
| 1.5152 | 18.0 | 88848 | nan |
|
| 67 |
+
| 1.5066 | 19.0 | 93784 | 1.5964 |
|
| 68 |
+
| 1.4764 | 20.0 | 98720 | 1.5965 |
|
| 69 |
+
|
| 70 |
+
|
| 71 |
+
### Framework versions
|
| 72 |
+
|
| 73 |
+
- Transformers 4.38.2
|
| 74 |
+
- Pytorch 1.13.1+cu116
|
| 75 |
+
- Datasets 2.18.0
|
| 76 |
+
- Tokenizers 0.15.2
|
config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google-bert/bert-base-cased",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForMaskedLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"gradient_checkpointing": false,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"layer_norm_eps": 1e-12,
|
| 15 |
+
"max_position_embeddings": 512,
|
| 16 |
+
"model_type": "bert",
|
| 17 |
+
"num_attention_heads": 12,
|
| 18 |
+
"num_hidden_layers": 12,
|
| 19 |
+
"pad_token_id": 0,
|
| 20 |
+
"position_embedding_type": "absolute",
|
| 21 |
+
"torch_dtype": "float32",
|
| 22 |
+
"transformers_version": "4.38.2",
|
| 23 |
+
"type_vocab_size": 2,
|
| 24 |
+
"use_cache": true,
|
| 25 |
+
"vocab_size": 28996
|
| 26 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"pad_token_id": 0,
|
| 4 |
+
"transformers_version": "4.38.2"
|
| 5 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6ea802980df924fe7ea3b4c5c293932c490f8d6edfe4235a9fa53f6ffdb32178
|
| 3 |
+
size 433386912
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1569 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 20.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 98720,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.1,
|
| 13 |
+
"grad_norm": 27.71003532409668,
|
| 14 |
+
"learning_rate": 9.949351701782822e-06,
|
| 15 |
+
"loss": 2.7967,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.2,
|
| 20 |
+
"grad_norm": 23.26209831237793,
|
| 21 |
+
"learning_rate": 9.898703403565641e-06,
|
| 22 |
+
"loss": 2.5399,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.3,
|
| 27 |
+
"grad_norm": 23.482177734375,
|
| 28 |
+
"learning_rate": 9.84805510534846e-06,
|
| 29 |
+
"loss": 2.5093,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.41,
|
| 34 |
+
"grad_norm": 22.170177459716797,
|
| 35 |
+
"learning_rate": 9.797406807131282e-06,
|
| 36 |
+
"loss": 2.4094,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.51,
|
| 41 |
+
"grad_norm": 19.412521362304688,
|
| 42 |
+
"learning_rate": 9.746758508914101e-06,
|
| 43 |
+
"loss": 2.3084,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.61,
|
| 48 |
+
"grad_norm": 28.102920532226562,
|
| 49 |
+
"learning_rate": 9.696110210696922e-06,
|
| 50 |
+
"loss": 2.364,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.71,
|
| 55 |
+
"grad_norm": 24.38800048828125,
|
| 56 |
+
"learning_rate": 9.645461912479742e-06,
|
| 57 |
+
"loss": 2.2956,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.81,
|
| 62 |
+
"grad_norm": 18.825523376464844,
|
| 63 |
+
"learning_rate": 9.594813614262561e-06,
|
| 64 |
+
"loss": 2.2741,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.91,
|
| 69 |
+
"grad_norm": 19.316570281982422,
|
| 70 |
+
"learning_rate": 9.544165316045382e-06,
|
| 71 |
+
"loss": 2.2161,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 1.0,
|
| 76 |
+
"eval_loss": NaN,
|
| 77 |
+
"eval_runtime": 77.6479,
|
| 78 |
+
"eval_samples_per_second": 127.151,
|
| 79 |
+
"eval_steps_per_second": 15.905,
|
| 80 |
+
"step": 4936
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.01,
|
| 84 |
+
"grad_norm": 17.73604393005371,
|
| 85 |
+
"learning_rate": 9.493517017828202e-06,
|
| 86 |
+
"loss": 2.2035,
|
| 87 |
+
"step": 5000
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.11,
|
| 91 |
+
"grad_norm": 24.61719512939453,
|
| 92 |
+
"learning_rate": 9.442868719611021e-06,
|
| 93 |
+
"loss": 2.1512,
|
| 94 |
+
"step": 5500
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 1.22,
|
| 98 |
+
"grad_norm": 24.99042510986328,
|
| 99 |
+
"learning_rate": 9.392220421393842e-06,
|
| 100 |
+
"loss": 2.1579,
|
| 101 |
+
"step": 6000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.32,
|
| 105 |
+
"grad_norm": 28.80664825439453,
|
| 106 |
+
"learning_rate": 9.341572123176662e-06,
|
| 107 |
+
"loss": 2.1624,
|
| 108 |
+
"step": 6500
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.42,
|
| 112 |
+
"grad_norm": 17.5797176361084,
|
| 113 |
+
"learning_rate": 9.290923824959483e-06,
|
| 114 |
+
"loss": 2.1107,
|
| 115 |
+
"step": 7000
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 1.52,
|
| 119 |
+
"grad_norm": 16.641563415527344,
|
| 120 |
+
"learning_rate": 9.240275526742302e-06,
|
| 121 |
+
"loss": 2.0927,
|
| 122 |
+
"step": 7500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 1.62,
|
| 126 |
+
"grad_norm": 20.170042037963867,
|
| 127 |
+
"learning_rate": 9.189627228525122e-06,
|
| 128 |
+
"loss": 2.1255,
|
| 129 |
+
"step": 8000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.72,
|
| 133 |
+
"grad_norm": 19.44417381286621,
|
| 134 |
+
"learning_rate": 9.138978930307943e-06,
|
| 135 |
+
"loss": 2.0959,
|
| 136 |
+
"step": 8500
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 1.82,
|
| 140 |
+
"grad_norm": 15.98162841796875,
|
| 141 |
+
"learning_rate": 9.088330632090762e-06,
|
| 142 |
+
"loss": 2.107,
|
| 143 |
+
"step": 9000
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.92,
|
| 147 |
+
"grad_norm": 26.916950225830078,
|
| 148 |
+
"learning_rate": 9.037682333873583e-06,
|
| 149 |
+
"loss": 2.0531,
|
| 150 |
+
"step": 9500
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 2.0,
|
| 154 |
+
"eval_loss": 1.9590071439743042,
|
| 155 |
+
"eval_runtime": 77.6637,
|
| 156 |
+
"eval_samples_per_second": 127.125,
|
| 157 |
+
"eval_steps_per_second": 15.902,
|
| 158 |
+
"step": 9872
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 2.03,
|
| 162 |
+
"grad_norm": 16.49506187438965,
|
| 163 |
+
"learning_rate": 8.987034035656403e-06,
|
| 164 |
+
"loss": 2.0818,
|
| 165 |
+
"step": 10000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.13,
|
| 169 |
+
"grad_norm": 33.3337516784668,
|
| 170 |
+
"learning_rate": 8.936385737439222e-06,
|
| 171 |
+
"loss": 2.083,
|
| 172 |
+
"step": 10500
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 2.23,
|
| 176 |
+
"grad_norm": 25.783584594726562,
|
| 177 |
+
"learning_rate": 8.885737439222043e-06,
|
| 178 |
+
"loss": 2.0203,
|
| 179 |
+
"step": 11000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 2.33,
|
| 183 |
+
"grad_norm": 15.96865177154541,
|
| 184 |
+
"learning_rate": 8.835089141004863e-06,
|
| 185 |
+
"loss": 1.9965,
|
| 186 |
+
"step": 11500
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 2.43,
|
| 190 |
+
"grad_norm": 17.585542678833008,
|
| 191 |
+
"learning_rate": 8.784440842787682e-06,
|
| 192 |
+
"loss": 2.0138,
|
| 193 |
+
"step": 12000
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 2.53,
|
| 197 |
+
"grad_norm": 17.34233856201172,
|
| 198 |
+
"learning_rate": 8.733792544570503e-06,
|
| 199 |
+
"loss": 1.9905,
|
| 200 |
+
"step": 12500
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 2.63,
|
| 204 |
+
"grad_norm": 22.320497512817383,
|
| 205 |
+
"learning_rate": 8.683144246353323e-06,
|
| 206 |
+
"loss": 2.0074,
|
| 207 |
+
"step": 13000
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 2.74,
|
| 211 |
+
"grad_norm": 19.803224563598633,
|
| 212 |
+
"learning_rate": 8.632495948136144e-06,
|
| 213 |
+
"loss": 1.9483,
|
| 214 |
+
"step": 13500
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 2.84,
|
| 218 |
+
"grad_norm": 17.85085678100586,
|
| 219 |
+
"learning_rate": 8.581847649918963e-06,
|
| 220 |
+
"loss": 2.014,
|
| 221 |
+
"step": 14000
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 2.94,
|
| 225 |
+
"grad_norm": 23.786117553710938,
|
| 226 |
+
"learning_rate": 8.531199351701782e-06,
|
| 227 |
+
"loss": 1.978,
|
| 228 |
+
"step": 14500
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 3.0,
|
| 232 |
+
"eval_loss": NaN,
|
| 233 |
+
"eval_runtime": 77.6234,
|
| 234 |
+
"eval_samples_per_second": 127.191,
|
| 235 |
+
"eval_steps_per_second": 15.91,
|
| 236 |
+
"step": 14808
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 3.04,
|
| 240 |
+
"grad_norm": 14.98945140838623,
|
| 241 |
+
"learning_rate": 8.480551053484604e-06,
|
| 242 |
+
"loss": 1.9568,
|
| 243 |
+
"step": 15000
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 3.14,
|
| 247 |
+
"grad_norm": 21.061824798583984,
|
| 248 |
+
"learning_rate": 8.429902755267425e-06,
|
| 249 |
+
"loss": 1.9454,
|
| 250 |
+
"step": 15500
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 3.24,
|
| 254 |
+
"grad_norm": 17.81777000427246,
|
| 255 |
+
"learning_rate": 8.379254457050244e-06,
|
| 256 |
+
"loss": 1.9033,
|
| 257 |
+
"step": 16000
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 3.34,
|
| 261 |
+
"grad_norm": 19.77060890197754,
|
| 262 |
+
"learning_rate": 8.328606158833064e-06,
|
| 263 |
+
"loss": 1.971,
|
| 264 |
+
"step": 16500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 3.44,
|
| 268 |
+
"grad_norm": 23.06456756591797,
|
| 269 |
+
"learning_rate": 8.277957860615885e-06,
|
| 270 |
+
"loss": 1.9344,
|
| 271 |
+
"step": 17000
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 3.55,
|
| 275 |
+
"grad_norm": 21.723846435546875,
|
| 276 |
+
"learning_rate": 8.227309562398704e-06,
|
| 277 |
+
"loss": 1.9551,
|
| 278 |
+
"step": 17500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 3.65,
|
| 282 |
+
"grad_norm": 14.961769104003906,
|
| 283 |
+
"learning_rate": 8.176661264181525e-06,
|
| 284 |
+
"loss": 1.8939,
|
| 285 |
+
"step": 18000
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 3.75,
|
| 289 |
+
"grad_norm": 35.004859924316406,
|
| 290 |
+
"learning_rate": 8.126012965964345e-06,
|
| 291 |
+
"loss": 1.9347,
|
| 292 |
+
"step": 18500
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 3.85,
|
| 296 |
+
"grad_norm": 20.03676986694336,
|
| 297 |
+
"learning_rate": 8.075364667747164e-06,
|
| 298 |
+
"loss": 1.8949,
|
| 299 |
+
"step": 19000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 3.95,
|
| 303 |
+
"grad_norm": 23.595853805541992,
|
| 304 |
+
"learning_rate": 8.024716369529985e-06,
|
| 305 |
+
"loss": 1.9616,
|
| 306 |
+
"step": 19500
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 4.0,
|
| 310 |
+
"eval_loss": NaN,
|
| 311 |
+
"eval_runtime": 77.6621,
|
| 312 |
+
"eval_samples_per_second": 127.128,
|
| 313 |
+
"eval_steps_per_second": 15.902,
|
| 314 |
+
"step": 19744
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 4.05,
|
| 318 |
+
"grad_norm": 24.895097732543945,
|
| 319 |
+
"learning_rate": 7.974068071312805e-06,
|
| 320 |
+
"loss": 1.9181,
|
| 321 |
+
"step": 20000
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 4.15,
|
| 325 |
+
"grad_norm": 16.272125244140625,
|
| 326 |
+
"learning_rate": 7.923419773095624e-06,
|
| 327 |
+
"loss": 1.8546,
|
| 328 |
+
"step": 20500
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 4.25,
|
| 332 |
+
"grad_norm": 16.912342071533203,
|
| 333 |
+
"learning_rate": 7.872771474878445e-06,
|
| 334 |
+
"loss": 1.8722,
|
| 335 |
+
"step": 21000
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 4.36,
|
| 339 |
+
"grad_norm": 19.567813873291016,
|
| 340 |
+
"learning_rate": 7.822123176661264e-06,
|
| 341 |
+
"loss": 1.8421,
|
| 342 |
+
"step": 21500
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 4.46,
|
| 346 |
+
"grad_norm": 24.266704559326172,
|
| 347 |
+
"learning_rate": 7.771474878444086e-06,
|
| 348 |
+
"loss": 1.8481,
|
| 349 |
+
"step": 22000
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 4.56,
|
| 353 |
+
"grad_norm": 39.59248733520508,
|
| 354 |
+
"learning_rate": 7.720826580226905e-06,
|
| 355 |
+
"loss": 1.8366,
|
| 356 |
+
"step": 22500
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 4.66,
|
| 360 |
+
"grad_norm": 28.474586486816406,
|
| 361 |
+
"learning_rate": 7.670178282009724e-06,
|
| 362 |
+
"loss": 1.8959,
|
| 363 |
+
"step": 23000
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 4.76,
|
| 367 |
+
"grad_norm": 25.967967987060547,
|
| 368 |
+
"learning_rate": 7.6195299837925455e-06,
|
| 369 |
+
"loss": 1.8536,
|
| 370 |
+
"step": 23500
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 4.86,
|
| 374 |
+
"grad_norm": 22.55752944946289,
|
| 375 |
+
"learning_rate": 7.568881685575366e-06,
|
| 376 |
+
"loss": 1.8809,
|
| 377 |
+
"step": 24000
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 4.96,
|
| 381 |
+
"grad_norm": 26.150907516479492,
|
| 382 |
+
"learning_rate": 7.518233387358186e-06,
|
| 383 |
+
"loss": 1.8559,
|
| 384 |
+
"step": 24500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 5.0,
|
| 388 |
+
"eval_loss": 1.8337851762771606,
|
| 389 |
+
"eval_runtime": 77.7133,
|
| 390 |
+
"eval_samples_per_second": 127.044,
|
| 391 |
+
"eval_steps_per_second": 15.892,
|
| 392 |
+
"step": 24680
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 5.06,
|
| 396 |
+
"grad_norm": 16.329139709472656,
|
| 397 |
+
"learning_rate": 7.4675850891410055e-06,
|
| 398 |
+
"loss": 1.855,
|
| 399 |
+
"step": 25000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 5.17,
|
| 403 |
+
"grad_norm": 21.760122299194336,
|
| 404 |
+
"learning_rate": 7.416936790923826e-06,
|
| 405 |
+
"loss": 1.8109,
|
| 406 |
+
"step": 25500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 5.27,
|
| 410 |
+
"grad_norm": 20.415075302124023,
|
| 411 |
+
"learning_rate": 7.366288492706646e-06,
|
| 412 |
+
"loss": 1.7785,
|
| 413 |
+
"step": 26000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 5.37,
|
| 417 |
+
"grad_norm": 14.889837265014648,
|
| 418 |
+
"learning_rate": 7.315640194489466e-06,
|
| 419 |
+
"loss": 1.8091,
|
| 420 |
+
"step": 26500
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 5.47,
|
| 424 |
+
"grad_norm": 20.167137145996094,
|
| 425 |
+
"learning_rate": 7.264991896272286e-06,
|
| 426 |
+
"loss": 1.7611,
|
| 427 |
+
"step": 27000
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 5.57,
|
| 431 |
+
"grad_norm": 23.077468872070312,
|
| 432 |
+
"learning_rate": 7.214343598055106e-06,
|
| 433 |
+
"loss": 1.817,
|
| 434 |
+
"step": 27500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 5.67,
|
| 438 |
+
"grad_norm": 21.367464065551758,
|
| 439 |
+
"learning_rate": 7.163695299837926e-06,
|
| 440 |
+
"loss": 1.7946,
|
| 441 |
+
"step": 28000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 5.77,
|
| 445 |
+
"grad_norm": 15.969755172729492,
|
| 446 |
+
"learning_rate": 7.1130470016207465e-06,
|
| 447 |
+
"loss": 1.7874,
|
| 448 |
+
"step": 28500
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 5.88,
|
| 452 |
+
"grad_norm": 21.294042587280273,
|
| 453 |
+
"learning_rate": 7.062398703403566e-06,
|
| 454 |
+
"loss": 1.7853,
|
| 455 |
+
"step": 29000
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 5.98,
|
| 459 |
+
"grad_norm": 19.84623908996582,
|
| 460 |
+
"learning_rate": 7.011750405186386e-06,
|
| 461 |
+
"loss": 1.7902,
|
| 462 |
+
"step": 29500
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 6.0,
|
| 466 |
+
"eval_loss": 1.7772495746612549,
|
| 467 |
+
"eval_runtime": 77.7233,
|
| 468 |
+
"eval_samples_per_second": 127.028,
|
| 469 |
+
"eval_steps_per_second": 15.89,
|
| 470 |
+
"step": 29616
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 6.08,
|
| 474 |
+
"grad_norm": 21.59821128845215,
|
| 475 |
+
"learning_rate": 6.961102106969206e-06,
|
| 476 |
+
"loss": 1.7444,
|
| 477 |
+
"step": 30000
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 6.18,
|
| 481 |
+
"grad_norm": 20.259178161621094,
|
| 482 |
+
"learning_rate": 6.910453808752027e-06,
|
| 483 |
+
"loss": 1.7723,
|
| 484 |
+
"step": 30500
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 6.28,
|
| 488 |
+
"grad_norm": 19.63451385498047,
|
| 489 |
+
"learning_rate": 6.859805510534847e-06,
|
| 490 |
+
"loss": 1.7154,
|
| 491 |
+
"step": 31000
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 6.38,
|
| 495 |
+
"grad_norm": 24.970842361450195,
|
| 496 |
+
"learning_rate": 6.809157212317666e-06,
|
| 497 |
+
"loss": 1.7357,
|
| 498 |
+
"step": 31500
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 6.48,
|
| 502 |
+
"grad_norm": 20.864076614379883,
|
| 503 |
+
"learning_rate": 6.758508914100487e-06,
|
| 504 |
+
"loss": 1.7623,
|
| 505 |
+
"step": 32000
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 6.58,
|
| 509 |
+
"grad_norm": 23.677553176879883,
|
| 510 |
+
"learning_rate": 6.707860615883307e-06,
|
| 511 |
+
"loss": 1.7245,
|
| 512 |
+
"step": 32500
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 6.69,
|
| 516 |
+
"grad_norm": 21.432083129882812,
|
| 517 |
+
"learning_rate": 6.657212317666127e-06,
|
| 518 |
+
"loss": 1.7456,
|
| 519 |
+
"step": 33000
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 6.79,
|
| 523 |
+
"grad_norm": 27.3277587890625,
|
| 524 |
+
"learning_rate": 6.6065640194489465e-06,
|
| 525 |
+
"loss": 1.778,
|
| 526 |
+
"step": 33500
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 6.89,
|
| 530 |
+
"grad_norm": 19.86969566345215,
|
| 531 |
+
"learning_rate": 6.555915721231767e-06,
|
| 532 |
+
"loss": 1.7793,
|
| 533 |
+
"step": 34000
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 6.99,
|
| 537 |
+
"grad_norm": 22.414249420166016,
|
| 538 |
+
"learning_rate": 6.505267423014587e-06,
|
| 539 |
+
"loss": 1.7603,
|
| 540 |
+
"step": 34500
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 7.0,
|
| 544 |
+
"eval_loss": 1.7686270475387573,
|
| 545 |
+
"eval_runtime": 77.7088,
|
| 546 |
+
"eval_samples_per_second": 127.051,
|
| 547 |
+
"eval_steps_per_second": 15.893,
|
| 548 |
+
"step": 34552
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 7.09,
|
| 552 |
+
"grad_norm": 26.82056999206543,
|
| 553 |
+
"learning_rate": 6.454619124797407e-06,
|
| 554 |
+
"loss": 1.7195,
|
| 555 |
+
"step": 35000
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 7.19,
|
| 559 |
+
"grad_norm": 31.928022384643555,
|
| 560 |
+
"learning_rate": 6.403970826580227e-06,
|
| 561 |
+
"loss": 1.7568,
|
| 562 |
+
"step": 35500
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 7.29,
|
| 566 |
+
"grad_norm": 20.51543426513672,
|
| 567 |
+
"learning_rate": 6.353322528363047e-06,
|
| 568 |
+
"loss": 1.709,
|
| 569 |
+
"step": 36000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 7.39,
|
| 573 |
+
"grad_norm": 21.46121597290039,
|
| 574 |
+
"learning_rate": 6.302674230145867e-06,
|
| 575 |
+
"loss": 1.7127,
|
| 576 |
+
"step": 36500
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 7.5,
|
| 580 |
+
"grad_norm": 21.47946548461914,
|
| 581 |
+
"learning_rate": 6.252025931928688e-06,
|
| 582 |
+
"loss": 1.7442,
|
| 583 |
+
"step": 37000
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 7.6,
|
| 587 |
+
"grad_norm": 21.469694137573242,
|
| 588 |
+
"learning_rate": 6.201377633711507e-06,
|
| 589 |
+
"loss": 1.7433,
|
| 590 |
+
"step": 37500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 7.7,
|
| 594 |
+
"grad_norm": 17.47914695739746,
|
| 595 |
+
"learning_rate": 6.150729335494327e-06,
|
| 596 |
+
"loss": 1.7518,
|
| 597 |
+
"step": 38000
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 7.8,
|
| 601 |
+
"grad_norm": 17.23418617248535,
|
| 602 |
+
"learning_rate": 6.100081037277148e-06,
|
| 603 |
+
"loss": 1.7166,
|
| 604 |
+
"step": 38500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 7.9,
|
| 608 |
+
"grad_norm": 20.12972640991211,
|
| 609 |
+
"learning_rate": 6.0494327390599686e-06,
|
| 610 |
+
"loss": 1.769,
|
| 611 |
+
"step": 39000
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 8.0,
|
| 615 |
+
"eval_loss": 1.736777424812317,
|
| 616 |
+
"eval_runtime": 77.6891,
|
| 617 |
+
"eval_samples_per_second": 127.083,
|
| 618 |
+
"eval_steps_per_second": 15.897,
|
| 619 |
+
"step": 39488
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 8.0,
|
| 623 |
+
"grad_norm": 16.57518196105957,
|
| 624 |
+
"learning_rate": 5.998784440842789e-06,
|
| 625 |
+
"loss": 1.6586,
|
| 626 |
+
"step": 39500
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 8.1,
|
| 630 |
+
"grad_norm": 21.029388427734375,
|
| 631 |
+
"learning_rate": 5.948136142625608e-06,
|
| 632 |
+
"loss": 1.6599,
|
| 633 |
+
"step": 40000
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 8.21,
|
| 637 |
+
"grad_norm": 11.840949058532715,
|
| 638 |
+
"learning_rate": 5.8974878444084285e-06,
|
| 639 |
+
"loss": 1.6946,
|
| 640 |
+
"step": 40500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 8.31,
|
| 644 |
+
"grad_norm": 18.389739990234375,
|
| 645 |
+
"learning_rate": 5.846839546191249e-06,
|
| 646 |
+
"loss": 1.6685,
|
| 647 |
+
"step": 41000
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 8.41,
|
| 651 |
+
"grad_norm": 19.934383392333984,
|
| 652 |
+
"learning_rate": 5.796191247974069e-06,
|
| 653 |
+
"loss": 1.7214,
|
| 654 |
+
"step": 41500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 8.51,
|
| 658 |
+
"grad_norm": 22.178720474243164,
|
| 659 |
+
"learning_rate": 5.7455429497568885e-06,
|
| 660 |
+
"loss": 1.7002,
|
| 661 |
+
"step": 42000
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 8.61,
|
| 665 |
+
"grad_norm": 29.597888946533203,
|
| 666 |
+
"learning_rate": 5.694894651539709e-06,
|
| 667 |
+
"loss": 1.7304,
|
| 668 |
+
"step": 42500
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 8.71,
|
| 672 |
+
"grad_norm": 20.78500747680664,
|
| 673 |
+
"learning_rate": 5.644246353322529e-06,
|
| 674 |
+
"loss": 1.6568,
|
| 675 |
+
"step": 43000
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 8.81,
|
| 679 |
+
"grad_norm": 23.99928855895996,
|
| 680 |
+
"learning_rate": 5.593598055105349e-06,
|
| 681 |
+
"loss": 1.692,
|
| 682 |
+
"step": 43500
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 8.91,
|
| 686 |
+
"grad_norm": 28.18294334411621,
|
| 687 |
+
"learning_rate": 5.542949756888169e-06,
|
| 688 |
+
"loss": 1.6527,
|
| 689 |
+
"step": 44000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 9.0,
|
| 693 |
+
"eval_loss": NaN,
|
| 694 |
+
"eval_runtime": 77.6864,
|
| 695 |
+
"eval_samples_per_second": 127.088,
|
| 696 |
+
"eval_steps_per_second": 15.897,
|
| 697 |
+
"step": 44424
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 9.02,
|
| 701 |
+
"grad_norm": 16.98941993713379,
|
| 702 |
+
"learning_rate": 5.492301458670989e-06,
|
| 703 |
+
"loss": 1.6694,
|
| 704 |
+
"step": 44500
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 9.12,
|
| 708 |
+
"grad_norm": 25.71541976928711,
|
| 709 |
+
"learning_rate": 5.441653160453809e-06,
|
| 710 |
+
"loss": 1.6672,
|
| 711 |
+
"step": 45000
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 9.22,
|
| 715 |
+
"grad_norm": 18.719865798950195,
|
| 716 |
+
"learning_rate": 5.3910048622366294e-06,
|
| 717 |
+
"loss": 1.6557,
|
| 718 |
+
"step": 45500
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 9.32,
|
| 722 |
+
"grad_norm": 25.50741195678711,
|
| 723 |
+
"learning_rate": 5.34035656401945e-06,
|
| 724 |
+
"loss": 1.6223,
|
| 725 |
+
"step": 46000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 9.42,
|
| 729 |
+
"grad_norm": 19.657133102416992,
|
| 730 |
+
"learning_rate": 5.289708265802269e-06,
|
| 731 |
+
"loss": 1.6683,
|
| 732 |
+
"step": 46500
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 9.52,
|
| 736 |
+
"grad_norm": 16.309303283691406,
|
| 737 |
+
"learning_rate": 5.239059967585089e-06,
|
| 738 |
+
"loss": 1.6707,
|
| 739 |
+
"step": 47000
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 9.62,
|
| 743 |
+
"grad_norm": 16.309322357177734,
|
| 744 |
+
"learning_rate": 5.18841166936791e-06,
|
| 745 |
+
"loss": 1.6603,
|
| 746 |
+
"step": 47500
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 9.72,
|
| 750 |
+
"grad_norm": 22.41241455078125,
|
| 751 |
+
"learning_rate": 5.13776337115073e-06,
|
| 752 |
+
"loss": 1.6055,
|
| 753 |
+
"step": 48000
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 9.83,
|
| 757 |
+
"grad_norm": 17.34325408935547,
|
| 758 |
+
"learning_rate": 5.087115072933549e-06,
|
| 759 |
+
"loss": 1.673,
|
| 760 |
+
"step": 48500
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 9.93,
|
| 764 |
+
"grad_norm": 23.603504180908203,
|
| 765 |
+
"learning_rate": 5.03646677471637e-06,
|
| 766 |
+
"loss": 1.65,
|
| 767 |
+
"step": 49000
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 10.0,
|
| 771 |
+
"eval_loss": 1.688843846321106,
|
| 772 |
+
"eval_runtime": 77.6895,
|
| 773 |
+
"eval_samples_per_second": 127.083,
|
| 774 |
+
"eval_steps_per_second": 15.897,
|
| 775 |
+
"step": 49360
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 10.03,
|
| 779 |
+
"grad_norm": 24.20751190185547,
|
| 780 |
+
"learning_rate": 4.98581847649919e-06,
|
| 781 |
+
"loss": 1.6285,
|
| 782 |
+
"step": 49500
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 10.13,
|
| 786 |
+
"grad_norm": 15.967161178588867,
|
| 787 |
+
"learning_rate": 4.93517017828201e-06,
|
| 788 |
+
"loss": 1.608,
|
| 789 |
+
"step": 50000
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 10.23,
|
| 793 |
+
"grad_norm": 15.834975242614746,
|
| 794 |
+
"learning_rate": 4.88452188006483e-06,
|
| 795 |
+
"loss": 1.6181,
|
| 796 |
+
"step": 50500
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 10.33,
|
| 800 |
+
"grad_norm": 36.941707611083984,
|
| 801 |
+
"learning_rate": 4.83387358184765e-06,
|
| 802 |
+
"loss": 1.6119,
|
| 803 |
+
"step": 51000
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 10.43,
|
| 807 |
+
"grad_norm": 8.939834594726562,
|
| 808 |
+
"learning_rate": 4.78322528363047e-06,
|
| 809 |
+
"loss": 1.6556,
|
| 810 |
+
"step": 51500
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 10.53,
|
| 814 |
+
"grad_norm": 15.55808162689209,
|
| 815 |
+
"learning_rate": 4.73257698541329e-06,
|
| 816 |
+
"loss": 1.6024,
|
| 817 |
+
"step": 52000
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 10.64,
|
| 821 |
+
"grad_norm": 18.59736442565918,
|
| 822 |
+
"learning_rate": 4.6819286871961106e-06,
|
| 823 |
+
"loss": 1.6258,
|
| 824 |
+
"step": 52500
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 10.74,
|
| 828 |
+
"grad_norm": 18.915645599365234,
|
| 829 |
+
"learning_rate": 4.631280388978931e-06,
|
| 830 |
+
"loss": 1.6147,
|
| 831 |
+
"step": 53000
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 10.84,
|
| 835 |
+
"grad_norm": 21.615427017211914,
|
| 836 |
+
"learning_rate": 4.580632090761751e-06,
|
| 837 |
+
"loss": 1.6278,
|
| 838 |
+
"step": 53500
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 10.94,
|
| 842 |
+
"grad_norm": 16.90226936340332,
|
| 843 |
+
"learning_rate": 4.529983792544571e-06,
|
| 844 |
+
"loss": 1.6281,
|
| 845 |
+
"step": 54000
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 11.0,
|
| 849 |
+
"eval_loss": NaN,
|
| 850 |
+
"eval_runtime": 77.7252,
|
| 851 |
+
"eval_samples_per_second": 127.024,
|
| 852 |
+
"eval_steps_per_second": 15.889,
|
| 853 |
+
"step": 54296
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 11.04,
|
| 857 |
+
"grad_norm": 17.557144165039062,
|
| 858 |
+
"learning_rate": 4.479335494327391e-06,
|
| 859 |
+
"loss": 1.6239,
|
| 860 |
+
"step": 54500
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 11.14,
|
| 864 |
+
"grad_norm": 30.636253356933594,
|
| 865 |
+
"learning_rate": 4.428687196110211e-06,
|
| 866 |
+
"loss": 1.6089,
|
| 867 |
+
"step": 55000
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 11.24,
|
| 871 |
+
"grad_norm": 7.480132102966309,
|
| 872 |
+
"learning_rate": 4.378038897893031e-06,
|
| 873 |
+
"loss": 1.6014,
|
| 874 |
+
"step": 55500
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 11.35,
|
| 878 |
+
"grad_norm": 17.750568389892578,
|
| 879 |
+
"learning_rate": 4.3273905996758516e-06,
|
| 880 |
+
"loss": 1.6182,
|
| 881 |
+
"step": 56000
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 11.45,
|
| 885 |
+
"grad_norm": 21.50258445739746,
|
| 886 |
+
"learning_rate": 4.276742301458671e-06,
|
| 887 |
+
"loss": 1.6185,
|
| 888 |
+
"step": 56500
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 11.55,
|
| 892 |
+
"grad_norm": 12.455598831176758,
|
| 893 |
+
"learning_rate": 4.226094003241491e-06,
|
| 894 |
+
"loss": 1.5705,
|
| 895 |
+
"step": 57000
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 11.65,
|
| 899 |
+
"grad_norm": 21.358774185180664,
|
| 900 |
+
"learning_rate": 4.1754457050243115e-06,
|
| 901 |
+
"loss": 1.6046,
|
| 902 |
+
"step": 57500
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 11.75,
|
| 906 |
+
"grad_norm": 23.460098266601562,
|
| 907 |
+
"learning_rate": 4.124797406807132e-06,
|
| 908 |
+
"loss": 1.604,
|
| 909 |
+
"step": 58000
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 11.85,
|
| 913 |
+
"grad_norm": 24.99700355529785,
|
| 914 |
+
"learning_rate": 4.074149108589951e-06,
|
| 915 |
+
"loss": 1.6132,
|
| 916 |
+
"step": 58500
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 11.95,
|
| 920 |
+
"grad_norm": 20.90674591064453,
|
| 921 |
+
"learning_rate": 4.0235008103727714e-06,
|
| 922 |
+
"loss": 1.591,
|
| 923 |
+
"step": 59000
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 12.0,
|
| 927 |
+
"eval_loss": 1.6575616598129272,
|
| 928 |
+
"eval_runtime": 77.7204,
|
| 929 |
+
"eval_samples_per_second": 127.032,
|
| 930 |
+
"eval_steps_per_second": 15.89,
|
| 931 |
+
"step": 59232
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 12.05,
|
| 935 |
+
"grad_norm": 23.823829650878906,
|
| 936 |
+
"learning_rate": 3.972852512155592e-06,
|
| 937 |
+
"loss": 1.6094,
|
| 938 |
+
"step": 59500
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 12.16,
|
| 942 |
+
"grad_norm": 18.17566680908203,
|
| 943 |
+
"learning_rate": 3.922204213938412e-06,
|
| 944 |
+
"loss": 1.5851,
|
| 945 |
+
"step": 60000
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 12.26,
|
| 949 |
+
"grad_norm": 19.279071807861328,
|
| 950 |
+
"learning_rate": 3.871555915721232e-06,
|
| 951 |
+
"loss": 1.6302,
|
| 952 |
+
"step": 60500
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 12.36,
|
| 956 |
+
"grad_norm": 19.40210723876953,
|
| 957 |
+
"learning_rate": 3.8209076175040525e-06,
|
| 958 |
+
"loss": 1.5933,
|
| 959 |
+
"step": 61000
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 12.46,
|
| 963 |
+
"grad_norm": 20.0490665435791,
|
| 964 |
+
"learning_rate": 3.7702593192868723e-06,
|
| 965 |
+
"loss": 1.6217,
|
| 966 |
+
"step": 61500
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 12.56,
|
| 970 |
+
"grad_norm": 17.142908096313477,
|
| 971 |
+
"learning_rate": 3.719611021069692e-06,
|
| 972 |
+
"loss": 1.5791,
|
| 973 |
+
"step": 62000
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 12.66,
|
| 977 |
+
"grad_norm": 23.866859436035156,
|
| 978 |
+
"learning_rate": 3.668962722852513e-06,
|
| 979 |
+
"loss": 1.6127,
|
| 980 |
+
"step": 62500
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 12.76,
|
| 984 |
+
"grad_norm": 21.511199951171875,
|
| 985 |
+
"learning_rate": 3.6183144246353323e-06,
|
| 986 |
+
"loss": 1.5558,
|
| 987 |
+
"step": 63000
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 12.86,
|
| 991 |
+
"grad_norm": 20.330402374267578,
|
| 992 |
+
"learning_rate": 3.567666126418153e-06,
|
| 993 |
+
"loss": 1.5865,
|
| 994 |
+
"step": 63500
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 12.97,
|
| 998 |
+
"grad_norm": 17.255298614501953,
|
| 999 |
+
"learning_rate": 3.517017828200973e-06,
|
| 1000 |
+
"loss": 1.5508,
|
| 1001 |
+
"step": 64000
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 13.0,
|
| 1005 |
+
"eval_loss": NaN,
|
| 1006 |
+
"eval_runtime": 77.745,
|
| 1007 |
+
"eval_samples_per_second": 126.992,
|
| 1008 |
+
"eval_steps_per_second": 15.885,
|
| 1009 |
+
"step": 64168
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 13.07,
|
| 1013 |
+
"grad_norm": 15.916613578796387,
|
| 1014 |
+
"learning_rate": 3.466369529983793e-06,
|
| 1015 |
+
"loss": 1.5599,
|
| 1016 |
+
"step": 64500
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 13.17,
|
| 1020 |
+
"grad_norm": 12.248026847839355,
|
| 1021 |
+
"learning_rate": 3.415721231766613e-06,
|
| 1022 |
+
"loss": 1.5774,
|
| 1023 |
+
"step": 65000
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 13.27,
|
| 1027 |
+
"grad_norm": 20.037677764892578,
|
| 1028 |
+
"learning_rate": 3.365072933549433e-06,
|
| 1029 |
+
"loss": 1.5538,
|
| 1030 |
+
"step": 65500
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 13.37,
|
| 1034 |
+
"grad_norm": 20.426361083984375,
|
| 1035 |
+
"learning_rate": 3.314424635332253e-06,
|
| 1036 |
+
"loss": 1.5883,
|
| 1037 |
+
"step": 66000
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 13.47,
|
| 1041 |
+
"grad_norm": 32.5631103515625,
|
| 1042 |
+
"learning_rate": 3.2637763371150733e-06,
|
| 1043 |
+
"loss": 1.558,
|
| 1044 |
+
"step": 66500
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 13.57,
|
| 1048 |
+
"grad_norm": 25.788705825805664,
|
| 1049 |
+
"learning_rate": 3.213128038897893e-06,
|
| 1050 |
+
"loss": 1.571,
|
| 1051 |
+
"step": 67000
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 13.68,
|
| 1055 |
+
"grad_norm": 15.069945335388184,
|
| 1056 |
+
"learning_rate": 3.1624797406807134e-06,
|
| 1057 |
+
"loss": 1.5732,
|
| 1058 |
+
"step": 67500
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 13.78,
|
| 1062 |
+
"grad_norm": 13.779557228088379,
|
| 1063 |
+
"learning_rate": 3.1118314424635336e-06,
|
| 1064 |
+
"loss": 1.5425,
|
| 1065 |
+
"step": 68000
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 13.88,
|
| 1069 |
+
"grad_norm": 15.08449935913086,
|
| 1070 |
+
"learning_rate": 3.0611831442463535e-06,
|
| 1071 |
+
"loss": 1.5852,
|
| 1072 |
+
"step": 68500
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 13.98,
|
| 1076 |
+
"grad_norm": 32.31583023071289,
|
| 1077 |
+
"learning_rate": 3.0105348460291737e-06,
|
| 1078 |
+
"loss": 1.5929,
|
| 1079 |
+
"step": 69000
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 14.0,
|
| 1083 |
+
"eval_loss": 1.6150667667388916,
|
| 1084 |
+
"eval_runtime": 77.7004,
|
| 1085 |
+
"eval_samples_per_second": 127.065,
|
| 1086 |
+
"eval_steps_per_second": 15.894,
|
| 1087 |
+
"step": 69104
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 14.08,
|
| 1091 |
+
"grad_norm": 25.524412155151367,
|
| 1092 |
+
"learning_rate": 2.9598865478119936e-06,
|
| 1093 |
+
"loss": 1.5315,
|
| 1094 |
+
"step": 69500
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 14.18,
|
| 1098 |
+
"grad_norm": 19.731473922729492,
|
| 1099 |
+
"learning_rate": 2.9092382495948142e-06,
|
| 1100 |
+
"loss": 1.536,
|
| 1101 |
+
"step": 70000
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 14.28,
|
| 1105 |
+
"grad_norm": 21.87810516357422,
|
| 1106 |
+
"learning_rate": 2.8585899513776337e-06,
|
| 1107 |
+
"loss": 1.512,
|
| 1108 |
+
"step": 70500
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 14.38,
|
| 1112 |
+
"grad_norm": 28.01495933532715,
|
| 1113 |
+
"learning_rate": 2.8079416531604543e-06,
|
| 1114 |
+
"loss": 1.5568,
|
| 1115 |
+
"step": 71000
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 14.49,
|
| 1119 |
+
"grad_norm": 17.939329147338867,
|
| 1120 |
+
"learning_rate": 2.757293354943274e-06,
|
| 1121 |
+
"loss": 1.5332,
|
| 1122 |
+
"step": 71500
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 14.59,
|
| 1126 |
+
"grad_norm": 20.130300521850586,
|
| 1127 |
+
"learning_rate": 2.7066450567260944e-06,
|
| 1128 |
+
"loss": 1.5775,
|
| 1129 |
+
"step": 72000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 14.69,
|
| 1133 |
+
"grad_norm": 12.78222942352295,
|
| 1134 |
+
"learning_rate": 2.6559967585089143e-06,
|
| 1135 |
+
"loss": 1.5827,
|
| 1136 |
+
"step": 72500
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 14.79,
|
| 1140 |
+
"grad_norm": 15.022797584533691,
|
| 1141 |
+
"learning_rate": 2.6053484602917346e-06,
|
| 1142 |
+
"loss": 1.5574,
|
| 1143 |
+
"step": 73000
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 14.89,
|
| 1147 |
+
"grad_norm": 16.1165714263916,
|
| 1148 |
+
"learning_rate": 2.5547001620745544e-06,
|
| 1149 |
+
"loss": 1.5317,
|
| 1150 |
+
"step": 73500
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 14.99,
|
| 1154 |
+
"grad_norm": 19.746944427490234,
|
| 1155 |
+
"learning_rate": 2.5040518638573747e-06,
|
| 1156 |
+
"loss": 1.5504,
|
| 1157 |
+
"step": 74000
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 15.0,
|
| 1161 |
+
"eval_loss": NaN,
|
| 1162 |
+
"eval_runtime": 77.7514,
|
| 1163 |
+
"eval_samples_per_second": 126.982,
|
| 1164 |
+
"eval_steps_per_second": 15.884,
|
| 1165 |
+
"step": 74040
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 15.09,
|
| 1169 |
+
"grad_norm": 25.471561431884766,
|
| 1170 |
+
"learning_rate": 2.453403565640195e-06,
|
| 1171 |
+
"loss": 1.5746,
|
| 1172 |
+
"step": 74500
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 15.19,
|
| 1176 |
+
"grad_norm": 15.089065551757812,
|
| 1177 |
+
"learning_rate": 2.4027552674230148e-06,
|
| 1178 |
+
"loss": 1.5614,
|
| 1179 |
+
"step": 75000
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 15.3,
|
| 1183 |
+
"grad_norm": 24.640731811523438,
|
| 1184 |
+
"learning_rate": 2.352106969205835e-06,
|
| 1185 |
+
"loss": 1.5522,
|
| 1186 |
+
"step": 75500
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 15.4,
|
| 1190 |
+
"grad_norm": 17.83973503112793,
|
| 1191 |
+
"learning_rate": 2.301458670988655e-06,
|
| 1192 |
+
"loss": 1.4853,
|
| 1193 |
+
"step": 76000
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 15.5,
|
| 1197 |
+
"grad_norm": 26.84282875061035,
|
| 1198 |
+
"learning_rate": 2.250810372771475e-06,
|
| 1199 |
+
"loss": 1.4772,
|
| 1200 |
+
"step": 76500
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 15.6,
|
| 1204 |
+
"grad_norm": 17.549413681030273,
|
| 1205 |
+
"learning_rate": 2.200162074554295e-06,
|
| 1206 |
+
"loss": 1.4863,
|
| 1207 |
+
"step": 77000
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 15.7,
|
| 1211 |
+
"grad_norm": 30.209253311157227,
|
| 1212 |
+
"learning_rate": 2.1495137763371152e-06,
|
| 1213 |
+
"loss": 1.5432,
|
| 1214 |
+
"step": 77500
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 15.8,
|
| 1218 |
+
"grad_norm": 16.511491775512695,
|
| 1219 |
+
"learning_rate": 2.098865478119935e-06,
|
| 1220 |
+
"loss": 1.5001,
|
| 1221 |
+
"step": 78000
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 15.9,
|
| 1225 |
+
"grad_norm": 22.751419067382812,
|
| 1226 |
+
"learning_rate": 2.0482171799027553e-06,
|
| 1227 |
+
"loss": 1.5492,
|
| 1228 |
+
"step": 78500
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 16.0,
|
| 1232 |
+
"eval_loss": 1.6249719858169556,
|
| 1233 |
+
"eval_runtime": 77.7315,
|
| 1234 |
+
"eval_samples_per_second": 127.014,
|
| 1235 |
+
"eval_steps_per_second": 15.888,
|
| 1236 |
+
"step": 78976
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 16.0,
|
| 1240 |
+
"grad_norm": 20.166460037231445,
|
| 1241 |
+
"learning_rate": 1.9975688816855756e-06,
|
| 1242 |
+
"loss": 1.4967,
|
| 1243 |
+
"step": 79000
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 16.11,
|
| 1247 |
+
"grad_norm": 12.910844802856445,
|
| 1248 |
+
"learning_rate": 1.946920583468396e-06,
|
| 1249 |
+
"loss": 1.5459,
|
| 1250 |
+
"step": 79500
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 16.21,
|
| 1254 |
+
"grad_norm": 15.807537078857422,
|
| 1255 |
+
"learning_rate": 1.8962722852512159e-06,
|
| 1256 |
+
"loss": 1.4823,
|
| 1257 |
+
"step": 80000
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 16.31,
|
| 1261 |
+
"grad_norm": 15.588082313537598,
|
| 1262 |
+
"learning_rate": 1.845623987034036e-06,
|
| 1263 |
+
"loss": 1.5009,
|
| 1264 |
+
"step": 80500
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 16.41,
|
| 1268 |
+
"grad_norm": 19.26714324951172,
|
| 1269 |
+
"learning_rate": 1.794975688816856e-06,
|
| 1270 |
+
"loss": 1.5067,
|
| 1271 |
+
"step": 81000
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 16.51,
|
| 1275 |
+
"grad_norm": 13.45703411102295,
|
| 1276 |
+
"learning_rate": 1.744327390599676e-06,
|
| 1277 |
+
"loss": 1.5131,
|
| 1278 |
+
"step": 81500
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 16.61,
|
| 1282 |
+
"grad_norm": 19.007076263427734,
|
| 1283 |
+
"learning_rate": 1.693679092382496e-06,
|
| 1284 |
+
"loss": 1.5047,
|
| 1285 |
+
"step": 82000
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 16.71,
|
| 1289 |
+
"grad_norm": 19.805233001708984,
|
| 1290 |
+
"learning_rate": 1.6430307941653161e-06,
|
| 1291 |
+
"loss": 1.5133,
|
| 1292 |
+
"step": 82500
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 16.82,
|
| 1296 |
+
"grad_norm": 16.605981826782227,
|
| 1297 |
+
"learning_rate": 1.5923824959481362e-06,
|
| 1298 |
+
"loss": 1.5214,
|
| 1299 |
+
"step": 83000
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 16.92,
|
| 1303 |
+
"grad_norm": 18.25604820251465,
|
| 1304 |
+
"learning_rate": 1.5417341977309562e-06,
|
| 1305 |
+
"loss": 1.5488,
|
| 1306 |
+
"step": 83500
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 17.0,
|
| 1310 |
+
"eval_loss": 1.6310968399047852,
|
| 1311 |
+
"eval_runtime": 77.7353,
|
| 1312 |
+
"eval_samples_per_second": 127.008,
|
| 1313 |
+
"eval_steps_per_second": 15.887,
|
| 1314 |
+
"step": 83912
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 17.02,
|
| 1318 |
+
"grad_norm": 21.933780670166016,
|
| 1319 |
+
"learning_rate": 1.4910858995137765e-06,
|
| 1320 |
+
"loss": 1.4996,
|
| 1321 |
+
"step": 84000
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 17.12,
|
| 1325 |
+
"grad_norm": 14.489141464233398,
|
| 1326 |
+
"learning_rate": 1.4404376012965966e-06,
|
| 1327 |
+
"loss": 1.5357,
|
| 1328 |
+
"step": 84500
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 17.22,
|
| 1332 |
+
"grad_norm": 24.40933609008789,
|
| 1333 |
+
"learning_rate": 1.3897893030794166e-06,
|
| 1334 |
+
"loss": 1.4992,
|
| 1335 |
+
"step": 85000
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 17.32,
|
| 1339 |
+
"grad_norm": 22.278676986694336,
|
| 1340 |
+
"learning_rate": 1.3391410048622367e-06,
|
| 1341 |
+
"loss": 1.508,
|
| 1342 |
+
"step": 85500
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 17.42,
|
| 1346 |
+
"grad_norm": 14.33293342590332,
|
| 1347 |
+
"learning_rate": 1.2884927066450567e-06,
|
| 1348 |
+
"loss": 1.5416,
|
| 1349 |
+
"step": 86000
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 17.52,
|
| 1353 |
+
"grad_norm": 9.797683715820312,
|
| 1354 |
+
"learning_rate": 1.237844408427877e-06,
|
| 1355 |
+
"loss": 1.5316,
|
| 1356 |
+
"step": 86500
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 17.63,
|
| 1360 |
+
"grad_norm": 22.992408752441406,
|
| 1361 |
+
"learning_rate": 1.187196110210697e-06,
|
| 1362 |
+
"loss": 1.5616,
|
| 1363 |
+
"step": 87000
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 17.73,
|
| 1367 |
+
"grad_norm": 22.246749877929688,
|
| 1368 |
+
"learning_rate": 1.136547811993517e-06,
|
| 1369 |
+
"loss": 1.5187,
|
| 1370 |
+
"step": 87500
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 17.83,
|
| 1374 |
+
"grad_norm": 15.32419204711914,
|
| 1375 |
+
"learning_rate": 1.0858995137763371e-06,
|
| 1376 |
+
"loss": 1.519,
|
| 1377 |
+
"step": 88000
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 17.93,
|
| 1381 |
+
"grad_norm": 15.239709854125977,
|
| 1382 |
+
"learning_rate": 1.0352512155591574e-06,
|
| 1383 |
+
"loss": 1.5152,
|
| 1384 |
+
"step": 88500
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 18.0,
|
| 1388 |
+
"eval_loss": NaN,
|
| 1389 |
+
"eval_runtime": 77.7064,
|
| 1390 |
+
"eval_samples_per_second": 127.055,
|
| 1391 |
+
"eval_steps_per_second": 15.893,
|
| 1392 |
+
"step": 88848
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 18.03,
|
| 1396 |
+
"grad_norm": 17.708662033081055,
|
| 1397 |
+
"learning_rate": 9.846029173419774e-07,
|
| 1398 |
+
"loss": 1.5052,
|
| 1399 |
+
"step": 89000
|
| 1400 |
+
},
|
| 1401 |
+
{
|
| 1402 |
+
"epoch": 18.13,
|
| 1403 |
+
"grad_norm": 21.949464797973633,
|
| 1404 |
+
"learning_rate": 9.339546191247975e-07,
|
| 1405 |
+
"loss": 1.4853,
|
| 1406 |
+
"step": 89500
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 18.23,
|
| 1410 |
+
"grad_norm": 22.907304763793945,
|
| 1411 |
+
"learning_rate": 8.833063209076175e-07,
|
| 1412 |
+
"loss": 1.5041,
|
| 1413 |
+
"step": 90000
|
| 1414 |
+
},
|
| 1415 |
+
{
|
| 1416 |
+
"epoch": 18.33,
|
| 1417 |
+
"grad_norm": 16.836776733398438,
|
| 1418 |
+
"learning_rate": 8.326580226904377e-07,
|
| 1419 |
+
"loss": 1.5395,
|
| 1420 |
+
"step": 90500
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 18.44,
|
| 1424 |
+
"grad_norm": 24.573055267333984,
|
| 1425 |
+
"learning_rate": 7.820097244732577e-07,
|
| 1426 |
+
"loss": 1.5261,
|
| 1427 |
+
"step": 91000
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 18.54,
|
| 1431 |
+
"grad_norm": 16.58083152770996,
|
| 1432 |
+
"learning_rate": 7.313614262560778e-07,
|
| 1433 |
+
"loss": 1.501,
|
| 1434 |
+
"step": 91500
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 18.64,
|
| 1438 |
+
"grad_norm": 13.214485168457031,
|
| 1439 |
+
"learning_rate": 6.807131280388978e-07,
|
| 1440 |
+
"loss": 1.4723,
|
| 1441 |
+
"step": 92000
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 18.74,
|
| 1445 |
+
"grad_norm": 20.009634017944336,
|
| 1446 |
+
"learning_rate": 6.300648298217181e-07,
|
| 1447 |
+
"loss": 1.5101,
|
| 1448 |
+
"step": 92500
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 18.84,
|
| 1452 |
+
"grad_norm": 17.555011749267578,
|
| 1453 |
+
"learning_rate": 5.794165316045381e-07,
|
| 1454 |
+
"loss": 1.495,
|
| 1455 |
+
"step": 93000
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 18.94,
|
| 1459 |
+
"grad_norm": 23.471906661987305,
|
| 1460 |
+
"learning_rate": 5.287682333873582e-07,
|
| 1461 |
+
"loss": 1.5066,
|
| 1462 |
+
"step": 93500
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"epoch": 19.0,
|
| 1466 |
+
"eval_loss": 1.5963941812515259,
|
| 1467 |
+
"eval_runtime": 77.7338,
|
| 1468 |
+
"eval_samples_per_second": 127.01,
|
| 1469 |
+
"eval_steps_per_second": 15.888,
|
| 1470 |
+
"step": 93784
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"epoch": 19.04,
|
| 1474 |
+
"grad_norm": 23.58425521850586,
|
| 1475 |
+
"learning_rate": 4.781199351701784e-07,
|
| 1476 |
+
"loss": 1.4951,
|
| 1477 |
+
"step": 94000
|
| 1478 |
+
},
|
| 1479 |
+
{
|
| 1480 |
+
"epoch": 19.15,
|
| 1481 |
+
"grad_norm": 33.22106170654297,
|
| 1482 |
+
"learning_rate": 4.274716369529984e-07,
|
| 1483 |
+
"loss": 1.46,
|
| 1484 |
+
"step": 94500
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 19.25,
|
| 1488 |
+
"grad_norm": 19.294275283813477,
|
| 1489 |
+
"learning_rate": 3.768233387358185e-07,
|
| 1490 |
+
"loss": 1.5119,
|
| 1491 |
+
"step": 95000
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"epoch": 19.35,
|
| 1495 |
+
"grad_norm": 16.35405158996582,
|
| 1496 |
+
"learning_rate": 3.261750405186386e-07,
|
| 1497 |
+
"loss": 1.495,
|
| 1498 |
+
"step": 95500
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 19.45,
|
| 1502 |
+
"grad_norm": 12.868066787719727,
|
| 1503 |
+
"learning_rate": 2.755267423014587e-07,
|
| 1504 |
+
"loss": 1.507,
|
| 1505 |
+
"step": 96000
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 19.55,
|
| 1509 |
+
"grad_norm": 18.227985382080078,
|
| 1510 |
+
"learning_rate": 2.248784440842788e-07,
|
| 1511 |
+
"loss": 1.5176,
|
| 1512 |
+
"step": 96500
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 19.65,
|
| 1516 |
+
"grad_norm": 23.572263717651367,
|
| 1517 |
+
"learning_rate": 1.7423014586709888e-07,
|
| 1518 |
+
"loss": 1.5273,
|
| 1519 |
+
"step": 97000
|
| 1520 |
+
},
|
| 1521 |
+
{
|
| 1522 |
+
"epoch": 19.75,
|
| 1523 |
+
"grad_norm": 13.487857818603516,
|
| 1524 |
+
"learning_rate": 1.2358184764991896e-07,
|
| 1525 |
+
"loss": 1.514,
|
| 1526 |
+
"step": 97500
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 19.85,
|
| 1530 |
+
"grad_norm": 13.97512435913086,
|
| 1531 |
+
"learning_rate": 7.293354943273906e-08,
|
| 1532 |
+
"loss": 1.5456,
|
| 1533 |
+
"step": 98000
|
| 1534 |
+
},
|
| 1535 |
+
{
|
| 1536 |
+
"epoch": 19.96,
|
| 1537 |
+
"grad_norm": 21.878570556640625,
|
| 1538 |
+
"learning_rate": 2.228525121555916e-08,
|
| 1539 |
+
"loss": 1.4764,
|
| 1540 |
+
"step": 98500
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 20.0,
|
| 1544 |
+
"eval_loss": 1.5965242385864258,
|
| 1545 |
+
"eval_runtime": 77.7339,
|
| 1546 |
+
"eval_samples_per_second": 127.01,
|
| 1547 |
+
"eval_steps_per_second": 15.888,
|
| 1548 |
+
"step": 98720
|
| 1549 |
+
},
|
| 1550 |
+
{
|
| 1551 |
+
"epoch": 20.0,
|
| 1552 |
+
"step": 98720,
|
| 1553 |
+
"total_flos": 2.0786493066313728e+17,
|
| 1554 |
+
"train_loss": 1.721275230976519,
|
| 1555 |
+
"train_runtime": 20874.6418,
|
| 1556 |
+
"train_samples_per_second": 37.833,
|
| 1557 |
+
"train_steps_per_second": 4.729
|
| 1558 |
+
}
|
| 1559 |
+
],
|
| 1560 |
+
"logging_steps": 500,
|
| 1561 |
+
"max_steps": 98720,
|
| 1562 |
+
"num_input_tokens_seen": 0,
|
| 1563 |
+
"num_train_epochs": 20,
|
| 1564 |
+
"save_steps": 500,
|
| 1565 |
+
"total_flos": 2.0786493066313728e+17,
|
| 1566 |
+
"train_batch_size": 8,
|
| 1567 |
+
"trial_name": null,
|
| 1568 |
+
"trial_params": null
|
| 1569 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:97e91b3e0388343c1ca5d355d889e3d6c34031cdf90f9b95e8fbbd5cf04885f6
|
| 3 |
+
size 4475
|