Commit
•
8d684d0
1
Parent(s):
9f747cd
first commit
Browse files- all_results.json +14 -0
- config.json +46 -0
- eval_results.json +9 -0
- global_step22000/mp_rank_00_model_states.pt +3 -0
- global_step22000/zero_pp_rank_0_mp_rank_00_optim_states.pt +3 -0
- merges.txt +0 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +1 -0
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- train_results.json +8 -0
- trainer_state.json +1169 -0
- training_args.bin +3 -0
- vocab.json +0 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 500.0,
|
| 3 |
+
"eval_loss": 0.038116455078125,
|
| 4 |
+
"eval_runtime": 2.8548,
|
| 5 |
+
"eval_samples": 324,
|
| 6 |
+
"eval_samples_per_second": 113.495,
|
| 7 |
+
"eval_steps_per_second": 14.362,
|
| 8 |
+
"perplexity": 1.0388522054491718,
|
| 9 |
+
"train_loss": 0.09131514180790294,
|
| 10 |
+
"train_runtime": 19622.0994,
|
| 11 |
+
"train_samples": 704,
|
| 12 |
+
"train_samples_per_second": 17.939,
|
| 13 |
+
"train_steps_per_second": 1.121
|
| 14 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "distilgpt2",
|
| 3 |
+
"_num_labels": 1,
|
| 4 |
+
"activation_function": "gelu_new",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"GPT2LMHeadModel"
|
| 7 |
+
],
|
| 8 |
+
"attn_pdrop": 0.1,
|
| 9 |
+
"bos_token_id": 50256,
|
| 10 |
+
"embd_pdrop": 0.1,
|
| 11 |
+
"eos_token_id": 50256,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "LABEL_0"
|
| 14 |
+
},
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"label2id": {
|
| 17 |
+
"LABEL_0": 0
|
| 18 |
+
},
|
| 19 |
+
"layer_norm_epsilon": 1e-05,
|
| 20 |
+
"model_type": "gpt2",
|
| 21 |
+
"n_ctx": 1024,
|
| 22 |
+
"n_embd": 768,
|
| 23 |
+
"n_head": 12,
|
| 24 |
+
"n_inner": null,
|
| 25 |
+
"n_layer": 6,
|
| 26 |
+
"n_positions": 1024,
|
| 27 |
+
"reorder_and_upcast_attn": false,
|
| 28 |
+
"resid_pdrop": 0.1,
|
| 29 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 30 |
+
"scale_attn_weights": true,
|
| 31 |
+
"summary_activation": null,
|
| 32 |
+
"summary_first_dropout": 0.1,
|
| 33 |
+
"summary_proj_to_labels": true,
|
| 34 |
+
"summary_type": "cls_index",
|
| 35 |
+
"summary_use_proj": true,
|
| 36 |
+
"task_specific_params": {
|
| 37 |
+
"text-generation": {
|
| 38 |
+
"do_sample": true,
|
| 39 |
+
"max_length": 50
|
| 40 |
+
}
|
| 41 |
+
},
|
| 42 |
+
"torch_dtype": "float16",
|
| 43 |
+
"transformers_version": "4.13.0",
|
| 44 |
+
"use_cache": false,
|
| 45 |
+
"vocab_size": 50257
|
| 46 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 500.0,
|
| 3 |
+
"eval_loss": 0.038116455078125,
|
| 4 |
+
"eval_runtime": 2.8548,
|
| 5 |
+
"eval_samples": 324,
|
| 6 |
+
"eval_samples_per_second": 113.495,
|
| 7 |
+
"eval_steps_per_second": 14.362,
|
| 8 |
+
"perplexity": 1.0388522054491718
|
| 9 |
+
}
|
global_step22000/mp_rank_00_model_states.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:81270315831f9f420869b9454403122b8bee818ae4a36a2c9a9f189a24d6c71f
|
| 3 |
+
size 170135099
|
global_step22000/zero_pp_rank_0_mp_rank_00_optim_states.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:06a0ac178a30b83fdd68fa864036ef06dc96cd4c9ab087878f42bf9e53a60b71
|
| 3 |
+
size 982958359
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5901fcd35610b8efe54c8512d35aeaee69076091f34ad75d92e7fbaa839f3052
|
| 3 |
+
size 170133371
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "unk_token": "<|endoftext|>"}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "distilgpt2", "tokenizer_class": "GPT2Tokenizer"}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 500.0,
|
| 3 |
+
"train_loss": 0.09131514180790294,
|
| 4 |
+
"train_runtime": 19622.0994,
|
| 5 |
+
"train_samples": 704,
|
| 6 |
+
"train_samples_per_second": 17.939,
|
| 7 |
+
"train_steps_per_second": 1.121
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1169 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 500.0,
|
| 5 |
+
"global_step": 22000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 4.55,
|
| 12 |
+
"eval_loss": 1.7275390625,
|
| 13 |
+
"eval_runtime": 2.8452,
|
| 14 |
+
"eval_samples_per_second": 113.875,
|
| 15 |
+
"eval_steps_per_second": 14.41,
|
| 16 |
+
"step": 200
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 9.09,
|
| 20 |
+
"eval_loss": 1.052734375,
|
| 21 |
+
"eval_runtime": 2.8537,
|
| 22 |
+
"eval_samples_per_second": 113.538,
|
| 23 |
+
"eval_steps_per_second": 14.367,
|
| 24 |
+
"step": 400
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 11.36,
|
| 28 |
+
"learning_rate": 5e-05,
|
| 29 |
+
"loss": 1.9751,
|
| 30 |
+
"step": 500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 13.64,
|
| 34 |
+
"eval_loss": 0.64697265625,
|
| 35 |
+
"eval_runtime": 2.8455,
|
| 36 |
+
"eval_samples_per_second": 113.862,
|
| 37 |
+
"eval_steps_per_second": 14.409,
|
| 38 |
+
"step": 600
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 18.18,
|
| 42 |
+
"eval_loss": 0.3955078125,
|
| 43 |
+
"eval_runtime": 2.8586,
|
| 44 |
+
"eval_samples_per_second": 113.342,
|
| 45 |
+
"eval_steps_per_second": 14.343,
|
| 46 |
+
"step": 800
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 22.73,
|
| 50 |
+
"learning_rate": 5e-05,
|
| 51 |
+
"loss": 0.7515,
|
| 52 |
+
"step": 1000
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 22.73,
|
| 56 |
+
"eval_loss": 0.25244140625,
|
| 57 |
+
"eval_runtime": 2.8476,
|
| 58 |
+
"eval_samples_per_second": 113.78,
|
| 59 |
+
"eval_steps_per_second": 14.398,
|
| 60 |
+
"step": 1000
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 27.27,
|
| 64 |
+
"eval_loss": 0.167236328125,
|
| 65 |
+
"eval_runtime": 2.8473,
|
| 66 |
+
"eval_samples_per_second": 113.792,
|
| 67 |
+
"eval_steps_per_second": 14.4,
|
| 68 |
+
"step": 1200
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 31.82,
|
| 72 |
+
"eval_loss": 0.1171875,
|
| 73 |
+
"eval_runtime": 2.8502,
|
| 74 |
+
"eval_samples_per_second": 113.678,
|
| 75 |
+
"eval_steps_per_second": 14.385,
|
| 76 |
+
"step": 1400
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 34.09,
|
| 80 |
+
"learning_rate": 5e-05,
|
| 81 |
+
"loss": 0.3181,
|
| 82 |
+
"step": 1500
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 36.36,
|
| 86 |
+
"eval_loss": 0.089599609375,
|
| 87 |
+
"eval_runtime": 2.853,
|
| 88 |
+
"eval_samples_per_second": 113.564,
|
| 89 |
+
"eval_steps_per_second": 14.371,
|
| 90 |
+
"step": 1600
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 40.91,
|
| 94 |
+
"eval_loss": 0.07159423828125,
|
| 95 |
+
"eval_runtime": 2.8542,
|
| 96 |
+
"eval_samples_per_second": 113.519,
|
| 97 |
+
"eval_steps_per_second": 14.365,
|
| 98 |
+
"step": 1800
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 45.45,
|
| 102 |
+
"learning_rate": 5e-05,
|
| 103 |
+
"loss": 0.1654,
|
| 104 |
+
"step": 2000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 45.45,
|
| 108 |
+
"eval_loss": 0.06207275390625,
|
| 109 |
+
"eval_runtime": 2.8559,
|
| 110 |
+
"eval_samples_per_second": 113.449,
|
| 111 |
+
"eval_steps_per_second": 14.356,
|
| 112 |
+
"step": 2000
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 50.0,
|
| 116 |
+
"eval_loss": 0.053802490234375,
|
| 117 |
+
"eval_runtime": 2.8461,
|
| 118 |
+
"eval_samples_per_second": 113.841,
|
| 119 |
+
"eval_steps_per_second": 14.406,
|
| 120 |
+
"step": 2200
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 54.55,
|
| 124 |
+
"eval_loss": 0.049713134765625,
|
| 125 |
+
"eval_runtime": 2.8533,
|
| 126 |
+
"eval_samples_per_second": 113.551,
|
| 127 |
+
"eval_steps_per_second": 14.369,
|
| 128 |
+
"step": 2400
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 56.82,
|
| 132 |
+
"learning_rate": 5e-05,
|
| 133 |
+
"loss": 0.1059,
|
| 134 |
+
"step": 2500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 59.09,
|
| 138 |
+
"eval_loss": 0.047698974609375,
|
| 139 |
+
"eval_runtime": 2.8541,
|
| 140 |
+
"eval_samples_per_second": 113.523,
|
| 141 |
+
"eval_steps_per_second": 14.366,
|
| 142 |
+
"step": 2600
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 63.64,
|
| 146 |
+
"eval_loss": 0.04449462890625,
|
| 147 |
+
"eval_runtime": 2.8515,
|
| 148 |
+
"eval_samples_per_second": 113.625,
|
| 149 |
+
"eval_steps_per_second": 14.379,
|
| 150 |
+
"step": 2800
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 68.18,
|
| 154 |
+
"learning_rate": 5e-05,
|
| 155 |
+
"loss": 0.0771,
|
| 156 |
+
"step": 3000
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 68.18,
|
| 160 |
+
"eval_loss": 0.04400634765625,
|
| 161 |
+
"eval_runtime": 2.853,
|
| 162 |
+
"eval_samples_per_second": 113.565,
|
| 163 |
+
"eval_steps_per_second": 14.371,
|
| 164 |
+
"step": 3000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 72.73,
|
| 168 |
+
"eval_loss": 0.03985595703125,
|
| 169 |
+
"eval_runtime": 2.8483,
|
| 170 |
+
"eval_samples_per_second": 113.753,
|
| 171 |
+
"eval_steps_per_second": 14.395,
|
| 172 |
+
"step": 3200
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 77.27,
|
| 176 |
+
"eval_loss": 0.039398193359375,
|
| 177 |
+
"eval_runtime": 2.849,
|
| 178 |
+
"eval_samples_per_second": 113.725,
|
| 179 |
+
"eval_steps_per_second": 14.391,
|
| 180 |
+
"step": 3400
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 79.55,
|
| 184 |
+
"learning_rate": 5e-05,
|
| 185 |
+
"loss": 0.0604,
|
| 186 |
+
"step": 3500
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 81.82,
|
| 190 |
+
"eval_loss": 0.03973388671875,
|
| 191 |
+
"eval_runtime": 2.8542,
|
| 192 |
+
"eval_samples_per_second": 113.518,
|
| 193 |
+
"eval_steps_per_second": 14.365,
|
| 194 |
+
"step": 3600
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 86.36,
|
| 198 |
+
"eval_loss": 0.038177490234375,
|
| 199 |
+
"eval_runtime": 2.85,
|
| 200 |
+
"eval_samples_per_second": 113.682,
|
| 201 |
+
"eval_steps_per_second": 14.386,
|
| 202 |
+
"step": 3800
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 90.91,
|
| 206 |
+
"learning_rate": 5e-05,
|
| 207 |
+
"loss": 0.05,
|
| 208 |
+
"step": 4000
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 90.91,
|
| 212 |
+
"eval_loss": 0.037384033203125,
|
| 213 |
+
"eval_runtime": 2.8526,
|
| 214 |
+
"eval_samples_per_second": 113.581,
|
| 215 |
+
"eval_steps_per_second": 14.373,
|
| 216 |
+
"step": 4000
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 95.45,
|
| 220 |
+
"eval_loss": 0.037322998046875,
|
| 221 |
+
"eval_runtime": 2.8554,
|
| 222 |
+
"eval_samples_per_second": 113.468,
|
| 223 |
+
"eval_steps_per_second": 14.359,
|
| 224 |
+
"step": 4200
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 100.0,
|
| 228 |
+
"eval_loss": 0.035858154296875,
|
| 229 |
+
"eval_runtime": 2.8533,
|
| 230 |
+
"eval_samples_per_second": 113.555,
|
| 231 |
+
"eval_steps_per_second": 14.37,
|
| 232 |
+
"step": 4400
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 102.27,
|
| 236 |
+
"learning_rate": 5e-05,
|
| 237 |
+
"loss": 0.0423,
|
| 238 |
+
"step": 4500
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 104.55,
|
| 242 |
+
"eval_loss": 0.036163330078125,
|
| 243 |
+
"eval_runtime": 2.8573,
|
| 244 |
+
"eval_samples_per_second": 113.392,
|
| 245 |
+
"eval_steps_per_second": 14.349,
|
| 246 |
+
"step": 4600
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 109.09,
|
| 250 |
+
"eval_loss": 0.0374755859375,
|
| 251 |
+
"eval_runtime": 2.8521,
|
| 252 |
+
"eval_samples_per_second": 113.599,
|
| 253 |
+
"eval_steps_per_second": 14.375,
|
| 254 |
+
"step": 4800
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 113.64,
|
| 258 |
+
"learning_rate": 5e-05,
|
| 259 |
+
"loss": 0.0364,
|
| 260 |
+
"step": 5000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 113.64,
|
| 264 |
+
"eval_loss": 0.035125732421875,
|
| 265 |
+
"eval_runtime": 2.8541,
|
| 266 |
+
"eval_samples_per_second": 113.521,
|
| 267 |
+
"eval_steps_per_second": 14.365,
|
| 268 |
+
"step": 5000
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 118.18,
|
| 272 |
+
"eval_loss": 0.034515380859375,
|
| 273 |
+
"eval_runtime": 2.8551,
|
| 274 |
+
"eval_samples_per_second": 113.483,
|
| 275 |
+
"eval_steps_per_second": 14.361,
|
| 276 |
+
"step": 5200
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 122.73,
|
| 280 |
+
"eval_loss": 0.036224365234375,
|
| 281 |
+
"eval_runtime": 2.8548,
|
| 282 |
+
"eval_samples_per_second": 113.491,
|
| 283 |
+
"eval_steps_per_second": 14.362,
|
| 284 |
+
"step": 5400
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 125.0,
|
| 288 |
+
"learning_rate": 5e-05,
|
| 289 |
+
"loss": 0.0321,
|
| 290 |
+
"step": 5500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 127.27,
|
| 294 |
+
"eval_loss": 0.0335693359375,
|
| 295 |
+
"eval_runtime": 2.8567,
|
| 296 |
+
"eval_samples_per_second": 113.418,
|
| 297 |
+
"eval_steps_per_second": 14.352,
|
| 298 |
+
"step": 5600
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 131.82,
|
| 302 |
+
"eval_loss": 0.03485107421875,
|
| 303 |
+
"eval_runtime": 2.853,
|
| 304 |
+
"eval_samples_per_second": 113.566,
|
| 305 |
+
"eval_steps_per_second": 14.371,
|
| 306 |
+
"step": 5800
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 136.36,
|
| 310 |
+
"learning_rate": 5e-05,
|
| 311 |
+
"loss": 0.0285,
|
| 312 |
+
"step": 6000
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 136.36,
|
| 316 |
+
"eval_loss": 0.03436279296875,
|
| 317 |
+
"eval_runtime": 2.8569,
|
| 318 |
+
"eval_samples_per_second": 113.41,
|
| 319 |
+
"eval_steps_per_second": 14.351,
|
| 320 |
+
"step": 6000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 140.91,
|
| 324 |
+
"eval_loss": 0.03497314453125,
|
| 325 |
+
"eval_runtime": 2.8533,
|
| 326 |
+
"eval_samples_per_second": 113.554,
|
| 327 |
+
"eval_steps_per_second": 14.37,
|
| 328 |
+
"step": 6200
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 145.45,
|
| 332 |
+
"eval_loss": 0.034088134765625,
|
| 333 |
+
"eval_runtime": 2.8565,
|
| 334 |
+
"eval_samples_per_second": 113.424,
|
| 335 |
+
"eval_steps_per_second": 14.353,
|
| 336 |
+
"step": 6400
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 147.73,
|
| 340 |
+
"learning_rate": 5e-05,
|
| 341 |
+
"loss": 0.0257,
|
| 342 |
+
"step": 6500
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 150.0,
|
| 346 |
+
"eval_loss": 0.0341796875,
|
| 347 |
+
"eval_runtime": 2.8575,
|
| 348 |
+
"eval_samples_per_second": 113.385,
|
| 349 |
+
"eval_steps_per_second": 14.348,
|
| 350 |
+
"step": 6600
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 154.55,
|
| 354 |
+
"eval_loss": 0.035675048828125,
|
| 355 |
+
"eval_runtime": 2.8571,
|
| 356 |
+
"eval_samples_per_second": 113.402,
|
| 357 |
+
"eval_steps_per_second": 14.35,
|
| 358 |
+
"step": 6800
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 159.09,
|
| 362 |
+
"learning_rate": 5e-05,
|
| 363 |
+
"loss": 0.0235,
|
| 364 |
+
"step": 7000
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 159.09,
|
| 368 |
+
"eval_loss": 0.0352783203125,
|
| 369 |
+
"eval_runtime": 2.854,
|
| 370 |
+
"eval_samples_per_second": 113.526,
|
| 371 |
+
"eval_steps_per_second": 14.366,
|
| 372 |
+
"step": 7000
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 163.64,
|
| 376 |
+
"eval_loss": 0.034881591796875,
|
| 377 |
+
"eval_runtime": 2.8546,
|
| 378 |
+
"eval_samples_per_second": 113.502,
|
| 379 |
+
"eval_steps_per_second": 14.363,
|
| 380 |
+
"step": 7200
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 168.18,
|
| 384 |
+
"eval_loss": 0.035003662109375,
|
| 385 |
+
"eval_runtime": 2.8564,
|
| 386 |
+
"eval_samples_per_second": 113.428,
|
| 387 |
+
"eval_steps_per_second": 14.354,
|
| 388 |
+
"step": 7400
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 170.45,
|
| 392 |
+
"learning_rate": 5e-05,
|
| 393 |
+
"loss": 0.0216,
|
| 394 |
+
"step": 7500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 172.73,
|
| 398 |
+
"eval_loss": 0.034149169921875,
|
| 399 |
+
"eval_runtime": 2.8551,
|
| 400 |
+
"eval_samples_per_second": 113.483,
|
| 401 |
+
"eval_steps_per_second": 14.36,
|
| 402 |
+
"step": 7600
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 177.27,
|
| 406 |
+
"eval_loss": 0.035186767578125,
|
| 407 |
+
"eval_runtime": 2.8549,
|
| 408 |
+
"eval_samples_per_second": 113.49,
|
| 409 |
+
"eval_steps_per_second": 14.361,
|
| 410 |
+
"step": 7800
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 181.82,
|
| 414 |
+
"learning_rate": 5e-05,
|
| 415 |
+
"loss": 0.0199,
|
| 416 |
+
"step": 8000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 181.82,
|
| 420 |
+
"eval_loss": 0.034149169921875,
|
| 421 |
+
"eval_runtime": 2.8566,
|
| 422 |
+
"eval_samples_per_second": 113.422,
|
| 423 |
+
"eval_steps_per_second": 14.353,
|
| 424 |
+
"step": 8000
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 186.36,
|
| 428 |
+
"eval_loss": 0.035308837890625,
|
| 429 |
+
"eval_runtime": 2.8525,
|
| 430 |
+
"eval_samples_per_second": 113.584,
|
| 431 |
+
"eval_steps_per_second": 14.373,
|
| 432 |
+
"step": 8200
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 190.91,
|
| 436 |
+
"eval_loss": 0.03472900390625,
|
| 437 |
+
"eval_runtime": 2.8548,
|
| 438 |
+
"eval_samples_per_second": 113.494,
|
| 439 |
+
"eval_steps_per_second": 14.362,
|
| 440 |
+
"step": 8400
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 193.18,
|
| 444 |
+
"learning_rate": 5e-05,
|
| 445 |
+
"loss": 0.0184,
|
| 446 |
+
"step": 8500
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 195.45,
|
| 450 |
+
"eval_loss": 0.034210205078125,
|
| 451 |
+
"eval_runtime": 2.8577,
|
| 452 |
+
"eval_samples_per_second": 113.377,
|
| 453 |
+
"eval_steps_per_second": 14.347,
|
| 454 |
+
"step": 8600
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 200.0,
|
| 458 |
+
"eval_loss": 0.03448486328125,
|
| 459 |
+
"eval_runtime": 2.8551,
|
| 460 |
+
"eval_samples_per_second": 113.48,
|
| 461 |
+
"eval_steps_per_second": 14.36,
|
| 462 |
+
"step": 8800
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 204.55,
|
| 466 |
+
"learning_rate": 5e-05,
|
| 467 |
+
"loss": 0.0173,
|
| 468 |
+
"step": 9000
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 204.55,
|
| 472 |
+
"eval_loss": 0.034454345703125,
|
| 473 |
+
"eval_runtime": 2.8538,
|
| 474 |
+
"eval_samples_per_second": 113.533,
|
| 475 |
+
"eval_steps_per_second": 14.367,
|
| 476 |
+
"step": 9000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 209.09,
|
| 480 |
+
"eval_loss": 0.0343017578125,
|
| 481 |
+
"eval_runtime": 2.8484,
|
| 482 |
+
"eval_samples_per_second": 113.747,
|
| 483 |
+
"eval_steps_per_second": 14.394,
|
| 484 |
+
"step": 9200
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 213.64,
|
| 488 |
+
"eval_loss": 0.0355224609375,
|
| 489 |
+
"eval_runtime": 2.852,
|
| 490 |
+
"eval_samples_per_second": 113.606,
|
| 491 |
+
"eval_steps_per_second": 14.376,
|
| 492 |
+
"step": 9400
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 215.91,
|
| 496 |
+
"learning_rate": 5e-05,
|
| 497 |
+
"loss": 0.0162,
|
| 498 |
+
"step": 9500
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 218.18,
|
| 502 |
+
"eval_loss": 0.0350341796875,
|
| 503 |
+
"eval_runtime": 2.8527,
|
| 504 |
+
"eval_samples_per_second": 113.575,
|
| 505 |
+
"eval_steps_per_second": 14.372,
|
| 506 |
+
"step": 9600
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 222.73,
|
| 510 |
+
"eval_loss": 0.035247802734375,
|
| 511 |
+
"eval_runtime": 2.8538,
|
| 512 |
+
"eval_samples_per_second": 113.532,
|
| 513 |
+
"eval_steps_per_second": 14.367,
|
| 514 |
+
"step": 9800
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 227.27,
|
| 518 |
+
"learning_rate": 5e-05,
|
| 519 |
+
"loss": 0.0152,
|
| 520 |
+
"step": 10000
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 227.27,
|
| 524 |
+
"eval_loss": 0.0350341796875,
|
| 525 |
+
"eval_runtime": 2.856,
|
| 526 |
+
"eval_samples_per_second": 113.444,
|
| 527 |
+
"eval_steps_per_second": 14.356,
|
| 528 |
+
"step": 10000
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 231.82,
|
| 532 |
+
"eval_loss": 0.0350341796875,
|
| 533 |
+
"eval_runtime": 2.8472,
|
| 534 |
+
"eval_samples_per_second": 113.795,
|
| 535 |
+
"eval_steps_per_second": 14.4,
|
| 536 |
+
"step": 10200
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 236.36,
|
| 540 |
+
"eval_loss": 0.035400390625,
|
| 541 |
+
"eval_runtime": 2.8527,
|
| 542 |
+
"eval_samples_per_second": 113.575,
|
| 543 |
+
"eval_steps_per_second": 14.372,
|
| 544 |
+
"step": 10400
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 238.64,
|
| 548 |
+
"learning_rate": 5e-05,
|
| 549 |
+
"loss": 0.0145,
|
| 550 |
+
"step": 10500
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 240.91,
|
| 554 |
+
"eval_loss": 0.035125732421875,
|
| 555 |
+
"eval_runtime": 2.8574,
|
| 556 |
+
"eval_samples_per_second": 113.392,
|
| 557 |
+
"eval_steps_per_second": 14.349,
|
| 558 |
+
"step": 10600
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 245.45,
|
| 562 |
+
"eval_loss": 0.03546142578125,
|
| 563 |
+
"eval_runtime": 2.8478,
|
| 564 |
+
"eval_samples_per_second": 113.773,
|
| 565 |
+
"eval_steps_per_second": 14.397,
|
| 566 |
+
"step": 10800
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 250.0,
|
| 570 |
+
"learning_rate": 5e-05,
|
| 571 |
+
"loss": 0.0137,
|
| 572 |
+
"step": 11000
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 250.0,
|
| 576 |
+
"eval_loss": 0.035552978515625,
|
| 577 |
+
"eval_runtime": 2.8538,
|
| 578 |
+
"eval_samples_per_second": 113.533,
|
| 579 |
+
"eval_steps_per_second": 14.367,
|
| 580 |
+
"step": 11000
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 254.55,
|
| 584 |
+
"eval_loss": 0.035736083984375,
|
| 585 |
+
"eval_runtime": 2.848,
|
| 586 |
+
"eval_samples_per_second": 113.764,
|
| 587 |
+
"eval_steps_per_second": 14.396,
|
| 588 |
+
"step": 11200
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 259.09,
|
| 592 |
+
"eval_loss": 0.03582763671875,
|
| 593 |
+
"eval_runtime": 2.8521,
|
| 594 |
+
"eval_samples_per_second": 113.601,
|
| 595 |
+
"eval_steps_per_second": 14.375,
|
| 596 |
+
"step": 11400
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 261.36,
|
| 600 |
+
"learning_rate": 5e-05,
|
| 601 |
+
"loss": 0.013,
|
| 602 |
+
"step": 11500
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 263.64,
|
| 606 |
+
"eval_loss": 0.03546142578125,
|
| 607 |
+
"eval_runtime": 2.8549,
|
| 608 |
+
"eval_samples_per_second": 113.489,
|
| 609 |
+
"eval_steps_per_second": 14.361,
|
| 610 |
+
"step": 11600
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 268.18,
|
| 614 |
+
"eval_loss": 0.035919189453125,
|
| 615 |
+
"eval_runtime": 2.8533,
|
| 616 |
+
"eval_samples_per_second": 113.553,
|
| 617 |
+
"eval_steps_per_second": 14.369,
|
| 618 |
+
"step": 11800
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 272.73,
|
| 622 |
+
"learning_rate": 5e-05,
|
| 623 |
+
"loss": 0.0127,
|
| 624 |
+
"step": 12000
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 272.73,
|
| 628 |
+
"eval_loss": 0.0361328125,
|
| 629 |
+
"eval_runtime": 2.8539,
|
| 630 |
+
"eval_samples_per_second": 113.529,
|
| 631 |
+
"eval_steps_per_second": 14.366,
|
| 632 |
+
"step": 12000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 277.27,
|
| 636 |
+
"eval_loss": 0.03619384765625,
|
| 637 |
+
"eval_runtime": 2.8463,
|
| 638 |
+
"eval_samples_per_second": 113.832,
|
| 639 |
+
"eval_steps_per_second": 14.405,
|
| 640 |
+
"step": 12200
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 281.82,
|
| 644 |
+
"eval_loss": 0.0361328125,
|
| 645 |
+
"eval_runtime": 2.8549,
|
| 646 |
+
"eval_samples_per_second": 113.491,
|
| 647 |
+
"eval_steps_per_second": 14.362,
|
| 648 |
+
"step": 12400
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 284.09,
|
| 652 |
+
"learning_rate": 5e-05,
|
| 653 |
+
"loss": 0.0118,
|
| 654 |
+
"step": 12500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 286.36,
|
| 658 |
+
"eval_loss": 0.035736083984375,
|
| 659 |
+
"eval_runtime": 2.8573,
|
| 660 |
+
"eval_samples_per_second": 113.395,
|
| 661 |
+
"eval_steps_per_second": 14.349,
|
| 662 |
+
"step": 12600
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 290.91,
|
| 666 |
+
"eval_loss": 0.036529541015625,
|
| 667 |
+
"eval_runtime": 2.8571,
|
| 668 |
+
"eval_samples_per_second": 113.402,
|
| 669 |
+
"eval_steps_per_second": 14.35,
|
| 670 |
+
"step": 12800
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 295.45,
|
| 674 |
+
"learning_rate": 5e-05,
|
| 675 |
+
"loss": 0.0114,
|
| 676 |
+
"step": 13000
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 295.45,
|
| 680 |
+
"eval_loss": 0.0372314453125,
|
| 681 |
+
"eval_runtime": 2.8553,
|
| 682 |
+
"eval_samples_per_second": 113.473,
|
| 683 |
+
"eval_steps_per_second": 14.359,
|
| 684 |
+
"step": 13000
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 300.0,
|
| 688 |
+
"eval_loss": 0.037384033203125,
|
| 689 |
+
"eval_runtime": 2.8451,
|
| 690 |
+
"eval_samples_per_second": 113.881,
|
| 691 |
+
"eval_steps_per_second": 14.411,
|
| 692 |
+
"step": 13200
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 304.55,
|
| 696 |
+
"eval_loss": 0.0396728515625,
|
| 697 |
+
"eval_runtime": 2.8561,
|
| 698 |
+
"eval_samples_per_second": 113.44,
|
| 699 |
+
"eval_steps_per_second": 14.355,
|
| 700 |
+
"step": 13400
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 306.82,
|
| 704 |
+
"learning_rate": 5e-05,
|
| 705 |
+
"loss": 0.0106,
|
| 706 |
+
"step": 13500
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 309.09,
|
| 710 |
+
"eval_loss": 0.036529541015625,
|
| 711 |
+
"eval_runtime": 2.8552,
|
| 712 |
+
"eval_samples_per_second": 113.478,
|
| 713 |
+
"eval_steps_per_second": 14.36,
|
| 714 |
+
"step": 13600
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 313.64,
|
| 718 |
+
"eval_loss": 0.036346435546875,
|
| 719 |
+
"eval_runtime": 2.8561,
|
| 720 |
+
"eval_samples_per_second": 113.442,
|
| 721 |
+
"eval_steps_per_second": 14.355,
|
| 722 |
+
"step": 13800
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 318.18,
|
| 726 |
+
"learning_rate": 5e-05,
|
| 727 |
+
"loss": 0.01,
|
| 728 |
+
"step": 14000
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 318.18,
|
| 732 |
+
"eval_loss": 0.037811279296875,
|
| 733 |
+
"eval_runtime": 2.8552,
|
| 734 |
+
"eval_samples_per_second": 113.477,
|
| 735 |
+
"eval_steps_per_second": 14.36,
|
| 736 |
+
"step": 14000
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 322.73,
|
| 740 |
+
"eval_loss": 0.036865234375,
|
| 741 |
+
"eval_runtime": 2.8537,
|
| 742 |
+
"eval_samples_per_second": 113.537,
|
| 743 |
+
"eval_steps_per_second": 14.367,
|
| 744 |
+
"step": 14200
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 327.27,
|
| 748 |
+
"eval_loss": 0.037261962890625,
|
| 749 |
+
"eval_runtime": 2.8546,
|
| 750 |
+
"eval_samples_per_second": 113.503,
|
| 751 |
+
"eval_steps_per_second": 14.363,
|
| 752 |
+
"step": 14400
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 329.55,
|
| 756 |
+
"learning_rate": 5e-05,
|
| 757 |
+
"loss": 0.0096,
|
| 758 |
+
"step": 14500
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 331.82,
|
| 762 |
+
"eval_loss": 0.036590576171875,
|
| 763 |
+
"eval_runtime": 2.8583,
|
| 764 |
+
"eval_samples_per_second": 113.355,
|
| 765 |
+
"eval_steps_per_second": 14.344,
|
| 766 |
+
"step": 14600
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 336.36,
|
| 770 |
+
"eval_loss": 0.037109375,
|
| 771 |
+
"eval_runtime": 2.8515,
|
| 772 |
+
"eval_samples_per_second": 113.625,
|
| 773 |
+
"eval_steps_per_second": 14.378,
|
| 774 |
+
"step": 14800
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 340.91,
|
| 778 |
+
"learning_rate": 5e-05,
|
| 779 |
+
"loss": 0.0093,
|
| 780 |
+
"step": 15000
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 340.91,
|
| 784 |
+
"eval_loss": 0.037567138671875,
|
| 785 |
+
"eval_runtime": 2.8557,
|
| 786 |
+
"eval_samples_per_second": 113.459,
|
| 787 |
+
"eval_steps_per_second": 14.357,
|
| 788 |
+
"step": 15000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 345.45,
|
| 792 |
+
"eval_loss": 0.03717041015625,
|
| 793 |
+
"eval_runtime": 2.8575,
|
| 794 |
+
"eval_samples_per_second": 113.384,
|
| 795 |
+
"eval_steps_per_second": 14.348,
|
| 796 |
+
"step": 15200
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 350.0,
|
| 800 |
+
"eval_loss": 0.03729248046875,
|
| 801 |
+
"eval_runtime": 2.8457,
|
| 802 |
+
"eval_samples_per_second": 113.856,
|
| 803 |
+
"eval_steps_per_second": 14.408,
|
| 804 |
+
"step": 15400
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 352.27,
|
| 808 |
+
"learning_rate": 5e-05,
|
| 809 |
+
"loss": 0.0087,
|
| 810 |
+
"step": 15500
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 354.55,
|
| 814 |
+
"eval_loss": 0.037506103515625,
|
| 815 |
+
"eval_runtime": 2.8601,
|
| 816 |
+
"eval_samples_per_second": 113.284,
|
| 817 |
+
"eval_steps_per_second": 14.335,
|
| 818 |
+
"step": 15600
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 359.09,
|
| 822 |
+
"eval_loss": 0.037445068359375,
|
| 823 |
+
"eval_runtime": 2.8482,
|
| 824 |
+
"eval_samples_per_second": 113.755,
|
| 825 |
+
"eval_steps_per_second": 14.395,
|
| 826 |
+
"step": 15800
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 363.64,
|
| 830 |
+
"learning_rate": 5e-05,
|
| 831 |
+
"loss": 0.0086,
|
| 832 |
+
"step": 16000
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 363.64,
|
| 836 |
+
"eval_loss": 0.037841796875,
|
| 837 |
+
"eval_runtime": 2.8578,
|
| 838 |
+
"eval_samples_per_second": 113.373,
|
| 839 |
+
"eval_steps_per_second": 14.347,
|
| 840 |
+
"step": 16000
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 368.18,
|
| 844 |
+
"eval_loss": 0.03704833984375,
|
| 845 |
+
"eval_runtime": 2.8552,
|
| 846 |
+
"eval_samples_per_second": 113.477,
|
| 847 |
+
"eval_steps_per_second": 14.36,
|
| 848 |
+
"step": 16200
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 372.73,
|
| 852 |
+
"eval_loss": 0.0380859375,
|
| 853 |
+
"eval_runtime": 2.851,
|
| 854 |
+
"eval_samples_per_second": 113.644,
|
| 855 |
+
"eval_steps_per_second": 14.381,
|
| 856 |
+
"step": 16400
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 375.0,
|
| 860 |
+
"learning_rate": 5e-05,
|
| 861 |
+
"loss": 0.0084,
|
| 862 |
+
"step": 16500
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 377.27,
|
| 866 |
+
"eval_loss": 0.037200927734375,
|
| 867 |
+
"eval_runtime": 2.851,
|
| 868 |
+
"eval_samples_per_second": 113.644,
|
| 869 |
+
"eval_steps_per_second": 14.381,
|
| 870 |
+
"step": 16600
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 381.82,
|
| 874 |
+
"eval_loss": 0.03680419921875,
|
| 875 |
+
"eval_runtime": 2.8529,
|
| 876 |
+
"eval_samples_per_second": 113.567,
|
| 877 |
+
"eval_steps_per_second": 14.371,
|
| 878 |
+
"step": 16800
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 386.36,
|
| 882 |
+
"learning_rate": 5e-05,
|
| 883 |
+
"loss": 0.0078,
|
| 884 |
+
"step": 17000
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 386.36,
|
| 888 |
+
"eval_loss": 0.036956787109375,
|
| 889 |
+
"eval_runtime": 2.8565,
|
| 890 |
+
"eval_samples_per_second": 113.425,
|
| 891 |
+
"eval_steps_per_second": 14.353,
|
| 892 |
+
"step": 17000
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 390.91,
|
| 896 |
+
"eval_loss": 0.03759765625,
|
| 897 |
+
"eval_runtime": 2.8479,
|
| 898 |
+
"eval_samples_per_second": 113.768,
|
| 899 |
+
"eval_steps_per_second": 14.397,
|
| 900 |
+
"step": 17200
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 395.45,
|
| 904 |
+
"eval_loss": 0.038238525390625,
|
| 905 |
+
"eval_runtime": 2.8538,
|
| 906 |
+
"eval_samples_per_second": 113.534,
|
| 907 |
+
"eval_steps_per_second": 14.367,
|
| 908 |
+
"step": 17400
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 397.73,
|
| 912 |
+
"learning_rate": 5e-05,
|
| 913 |
+
"loss": 0.0076,
|
| 914 |
+
"step": 17500
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 400.0,
|
| 918 |
+
"eval_loss": 0.03778076171875,
|
| 919 |
+
"eval_runtime": 2.8549,
|
| 920 |
+
"eval_samples_per_second": 113.488,
|
| 921 |
+
"eval_steps_per_second": 14.361,
|
| 922 |
+
"step": 17600
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 404.55,
|
| 926 |
+
"eval_loss": 0.038482666015625,
|
| 927 |
+
"eval_runtime": 2.8533,
|
| 928 |
+
"eval_samples_per_second": 113.554,
|
| 929 |
+
"eval_steps_per_second": 14.37,
|
| 930 |
+
"step": 17800
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 409.09,
|
| 934 |
+
"learning_rate": 5e-05,
|
| 935 |
+
"loss": 0.0073,
|
| 936 |
+
"step": 18000
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 409.09,
|
| 940 |
+
"eval_loss": 0.037872314453125,
|
| 941 |
+
"eval_runtime": 2.8518,
|
| 942 |
+
"eval_samples_per_second": 113.614,
|
| 943 |
+
"eval_steps_per_second": 14.377,
|
| 944 |
+
"step": 18000
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 413.64,
|
| 948 |
+
"eval_loss": 0.038787841796875,
|
| 949 |
+
"eval_runtime": 2.8524,
|
| 950 |
+
"eval_samples_per_second": 113.588,
|
| 951 |
+
"eval_steps_per_second": 14.374,
|
| 952 |
+
"step": 18200
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 418.18,
|
| 956 |
+
"eval_loss": 0.0379638671875,
|
| 957 |
+
"eval_runtime": 2.848,
|
| 958 |
+
"eval_samples_per_second": 113.766,
|
| 959 |
+
"eval_steps_per_second": 14.396,
|
| 960 |
+
"step": 18400
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 420.45,
|
| 964 |
+
"learning_rate": 5e-05,
|
| 965 |
+
"loss": 0.0072,
|
| 966 |
+
"step": 18500
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 422.73,
|
| 970 |
+
"eval_loss": 0.0380859375,
|
| 971 |
+
"eval_runtime": 2.8569,
|
| 972 |
+
"eval_samples_per_second": 113.409,
|
| 973 |
+
"eval_steps_per_second": 14.351,
|
| 974 |
+
"step": 18600
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 427.27,
|
| 978 |
+
"eval_loss": 0.037506103515625,
|
| 979 |
+
"eval_runtime": 2.8521,
|
| 980 |
+
"eval_samples_per_second": 113.599,
|
| 981 |
+
"eval_steps_per_second": 14.375,
|
| 982 |
+
"step": 18800
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 431.82,
|
| 986 |
+
"learning_rate": 5e-05,
|
| 987 |
+
"loss": 0.0072,
|
| 988 |
+
"step": 19000
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 431.82,
|
| 992 |
+
"eval_loss": 0.038818359375,
|
| 993 |
+
"eval_runtime": 2.8529,
|
| 994 |
+
"eval_samples_per_second": 113.569,
|
| 995 |
+
"eval_steps_per_second": 14.371,
|
| 996 |
+
"step": 19000
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 436.36,
|
| 1000 |
+
"eval_loss": 0.0380859375,
|
| 1001 |
+
"eval_runtime": 2.8546,
|
| 1002 |
+
"eval_samples_per_second": 113.501,
|
| 1003 |
+
"eval_steps_per_second": 14.363,
|
| 1004 |
+
"step": 19200
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 440.91,
|
| 1008 |
+
"eval_loss": 0.037322998046875,
|
| 1009 |
+
"eval_runtime": 2.8547,
|
| 1010 |
+
"eval_samples_per_second": 113.496,
|
| 1011 |
+
"eval_steps_per_second": 14.362,
|
| 1012 |
+
"step": 19400
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 443.18,
|
| 1016 |
+
"learning_rate": 5e-05,
|
| 1017 |
+
"loss": 0.0066,
|
| 1018 |
+
"step": 19500
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 445.45,
|
| 1022 |
+
"eval_loss": 0.037567138671875,
|
| 1023 |
+
"eval_runtime": 2.8553,
|
| 1024 |
+
"eval_samples_per_second": 113.474,
|
| 1025 |
+
"eval_steps_per_second": 14.359,
|
| 1026 |
+
"step": 19600
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 450.0,
|
| 1030 |
+
"eval_loss": 0.038421630859375,
|
| 1031 |
+
"eval_runtime": 2.8536,
|
| 1032 |
+
"eval_samples_per_second": 113.543,
|
| 1033 |
+
"eval_steps_per_second": 14.368,
|
| 1034 |
+
"step": 19800
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 454.55,
|
| 1038 |
+
"learning_rate": 5e-05,
|
| 1039 |
+
"loss": 0.0066,
|
| 1040 |
+
"step": 20000
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 454.55,
|
| 1044 |
+
"eval_loss": 0.0384521484375,
|
| 1045 |
+
"eval_runtime": 2.8565,
|
| 1046 |
+
"eval_samples_per_second": 113.424,
|
| 1047 |
+
"eval_steps_per_second": 14.353,
|
| 1048 |
+
"step": 20000
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 459.09,
|
| 1052 |
+
"eval_loss": 0.039764404296875,
|
| 1053 |
+
"eval_runtime": 2.858,
|
| 1054 |
+
"eval_samples_per_second": 113.365,
|
| 1055 |
+
"eval_steps_per_second": 14.346,
|
| 1056 |
+
"step": 20200
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 463.64,
|
| 1060 |
+
"eval_loss": 0.0384521484375,
|
| 1061 |
+
"eval_runtime": 2.8472,
|
| 1062 |
+
"eval_samples_per_second": 113.796,
|
| 1063 |
+
"eval_steps_per_second": 14.4,
|
| 1064 |
+
"step": 20400
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 465.91,
|
| 1068 |
+
"learning_rate": 5e-05,
|
| 1069 |
+
"loss": 0.0064,
|
| 1070 |
+
"step": 20500
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 468.18,
|
| 1074 |
+
"eval_loss": 0.038665771484375,
|
| 1075 |
+
"eval_runtime": 2.8586,
|
| 1076 |
+
"eval_samples_per_second": 113.343,
|
| 1077 |
+
"eval_steps_per_second": 14.343,
|
| 1078 |
+
"step": 20600
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 472.73,
|
| 1082 |
+
"eval_loss": 0.038970947265625,
|
| 1083 |
+
"eval_runtime": 2.8462,
|
| 1084 |
+
"eval_samples_per_second": 113.837,
|
| 1085 |
+
"eval_steps_per_second": 14.405,
|
| 1086 |
+
"step": 20800
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 477.27,
|
| 1090 |
+
"learning_rate": 5e-05,
|
| 1091 |
+
"loss": 0.0063,
|
| 1092 |
+
"step": 21000
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 477.27,
|
| 1096 |
+
"eval_loss": 0.037750244140625,
|
| 1097 |
+
"eval_runtime": 2.8576,
|
| 1098 |
+
"eval_samples_per_second": 113.383,
|
| 1099 |
+
"eval_steps_per_second": 14.348,
|
| 1100 |
+
"step": 21000
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 481.82,
|
| 1104 |
+
"eval_loss": 0.039154052734375,
|
| 1105 |
+
"eval_runtime": 2.8571,
|
| 1106 |
+
"eval_samples_per_second": 113.4,
|
| 1107 |
+
"eval_steps_per_second": 14.35,
|
| 1108 |
+
"step": 21200
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 486.36,
|
| 1112 |
+
"eval_loss": 0.037994384765625,
|
| 1113 |
+
"eval_runtime": 2.8502,
|
| 1114 |
+
"eval_samples_per_second": 113.674,
|
| 1115 |
+
"eval_steps_per_second": 14.385,
|
| 1116 |
+
"step": 21400
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 488.64,
|
| 1120 |
+
"learning_rate": 5e-05,
|
| 1121 |
+
"loss": 0.006,
|
| 1122 |
+
"step": 21500
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 490.91,
|
| 1126 |
+
"eval_loss": 0.0377197265625,
|
| 1127 |
+
"eval_runtime": 2.8583,
|
| 1128 |
+
"eval_samples_per_second": 113.356,
|
| 1129 |
+
"eval_steps_per_second": 14.344,
|
| 1130 |
+
"step": 21600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 495.45,
|
| 1134 |
+
"eval_loss": 0.03857421875,
|
| 1135 |
+
"eval_runtime": 2.8515,
|
| 1136 |
+
"eval_samples_per_second": 113.623,
|
| 1137 |
+
"eval_steps_per_second": 14.378,
|
| 1138 |
+
"step": 21800
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 500.0,
|
| 1142 |
+
"learning_rate": 5e-05,
|
| 1143 |
+
"loss": 0.006,
|
| 1144 |
+
"step": 22000
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 500.0,
|
| 1148 |
+
"eval_loss": 0.038116455078125,
|
| 1149 |
+
"eval_runtime": 2.8565,
|
| 1150 |
+
"eval_samples_per_second": 113.426,
|
| 1151 |
+
"eval_steps_per_second": 14.353,
|
| 1152 |
+
"step": 22000
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 500.0,
|
| 1156 |
+
"step": 22000,
|
| 1157 |
+
"total_flos": 9.1976456208384e+16,
|
| 1158 |
+
"train_loss": 0.09131514180790294,
|
| 1159 |
+
"train_runtime": 19622.0994,
|
| 1160 |
+
"train_samples_per_second": 17.939,
|
| 1161 |
+
"train_steps_per_second": 1.121
|
| 1162 |
+
}
|
| 1163 |
+
],
|
| 1164 |
+
"max_steps": 22000,
|
| 1165 |
+
"num_train_epochs": 500,
|
| 1166 |
+
"total_flos": 9.1976456208384e+16,
|
| 1167 |
+
"trial_name": null,
|
| 1168 |
+
"trial_params": null
|
| 1169 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:05d01b381fba99d5015de48297a40e3151d5f45edf8e62cb06835365185f1975
|
| 3 |
+
size 3887
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|