End of training
Browse files- all_results.json +8 -0
- preprocessor_config.json +28 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +58 -0
- train_results.json +8 -0
- trainer_state.json +735 -0
- vocab.txt +0 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.998542037907015,
|
| 3 |
+
"total_flos": 1.461291118888168e+18,
|
| 4 |
+
"train_loss": 1.202193304075487,
|
| 5 |
+
"train_runtime": 14493.7066,
|
| 6 |
+
"train_samples_per_second": 255.536,
|
| 7 |
+
"train_steps_per_second": 2.129
|
| 8 |
+
}
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"crop_size": {
|
| 3 |
+
"height": 224,
|
| 4 |
+
"width": 224
|
| 5 |
+
},
|
| 6 |
+
"do_center_crop": false,
|
| 7 |
+
"do_convert_rgb": true,
|
| 8 |
+
"do_normalize": true,
|
| 9 |
+
"do_rescale": true,
|
| 10 |
+
"do_resize": true,
|
| 11 |
+
"image_mean": [
|
| 12 |
+
0.48145466,
|
| 13 |
+
0.4578275,
|
| 14 |
+
0.40821073
|
| 15 |
+
],
|
| 16 |
+
"image_processor_type": "ChineseCLIPImageProcessor",
|
| 17 |
+
"image_std": [
|
| 18 |
+
0.26862954,
|
| 19 |
+
0.26130258,
|
| 20 |
+
0.27577711
|
| 21 |
+
],
|
| 22 |
+
"resample": 3,
|
| 23 |
+
"rescale_factor": 0.00392156862745098,
|
| 24 |
+
"size": {
|
| 25 |
+
"height": 224,
|
| 26 |
+
"width": 224
|
| 27 |
+
}
|
| 28 |
+
}
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"unk_token": "[UNK]"
|
| 7 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": true,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_basic_tokenize": true,
|
| 47 |
+
"do_lower_case": true,
|
| 48 |
+
"extra_special_tokens": {},
|
| 49 |
+
"mask_token": "[MASK]",
|
| 50 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 51 |
+
"never_split": null,
|
| 52 |
+
"pad_token": "[PAD]",
|
| 53 |
+
"sep_token": "[SEP]",
|
| 54 |
+
"strip_accents": null,
|
| 55 |
+
"tokenize_chinese_chars": true,
|
| 56 |
+
"tokenizer_class": "BertTokenizer",
|
| 57 |
+
"unk_token": "[UNK]"
|
| 58 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.998542037907015,
|
| 3 |
+
"total_flos": 1.461291118888168e+18,
|
| 4 |
+
"train_loss": 1.202193304075487,
|
| 5 |
+
"train_runtime": 14493.7066,
|
| 6 |
+
"train_samples_per_second": 255.536,
|
| 7 |
+
"train_steps_per_second": 2.129
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,735 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 9.998542037907015,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 30860,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.10011339705167666,
|
| 13 |
+
"grad_norm": 0.018291741609573364,
|
| 14 |
+
"learning_rate": 9.90051847051199e-05,
|
| 15 |
+
"loss": 2.1839,
|
| 16 |
+
"step": 309
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.2002267941033533,
|
| 20 |
+
"grad_norm": 0.02208893373608589,
|
| 21 |
+
"learning_rate": 9.800388852883993e-05,
|
| 22 |
+
"loss": 2.1361,
|
| 23 |
+
"step": 618
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.30034019115502997,
|
| 27 |
+
"grad_norm": 0.012775925919413567,
|
| 28 |
+
"learning_rate": 9.700259235255995e-05,
|
| 29 |
+
"loss": 2.1625,
|
| 30 |
+
"step": 927
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.4004535882067066,
|
| 34 |
+
"grad_norm": 0.016916805878281593,
|
| 35 |
+
"learning_rate": 9.600129617627997e-05,
|
| 36 |
+
"loss": 2.1534,
|
| 37 |
+
"step": 1236
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.5005669852583833,
|
| 41 |
+
"grad_norm": 0.006638580933213234,
|
| 42 |
+
"learning_rate": 9.5e-05,
|
| 43 |
+
"loss": 2.1464,
|
| 44 |
+
"step": 1545
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.6006803823100599,
|
| 48 |
+
"grad_norm": 0.013361390680074692,
|
| 49 |
+
"learning_rate": 9.399870382372004e-05,
|
| 50 |
+
"loss": 2.1685,
|
| 51 |
+
"step": 1854
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.7007937793617366,
|
| 55 |
+
"grad_norm": 0.005704471841454506,
|
| 56 |
+
"learning_rate": 9.300064808814e-05,
|
| 57 |
+
"loss": 2.1539,
|
| 58 |
+
"step": 2163
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.8009071764134132,
|
| 62 |
+
"grad_norm": 0.02282548137009144,
|
| 63 |
+
"learning_rate": 9.199935191186002e-05,
|
| 64 |
+
"loss": 2.1799,
|
| 65 |
+
"step": 2472
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.9010205734650899,
|
| 69 |
+
"grad_norm": 0.011914879083633423,
|
| 70 |
+
"learning_rate": 9.099805573558005e-05,
|
| 71 |
+
"loss": 2.1487,
|
| 72 |
+
"step": 2781
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 1.000971974728657,
|
| 76 |
+
"grad_norm": 0.020065952092409134,
|
| 77 |
+
"learning_rate": 8.999675955930007e-05,
|
| 78 |
+
"loss": 2.1416,
|
| 79 |
+
"step": 3090
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 1.1010853717803337,
|
| 83 |
+
"grad_norm": 0.020726900547742844,
|
| 84 |
+
"learning_rate": 8.899546338302009e-05,
|
| 85 |
+
"loss": 2.1422,
|
| 86 |
+
"step": 3399
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 1.2011987688320103,
|
| 90 |
+
"grad_norm": 0.02327028475701809,
|
| 91 |
+
"learning_rate": 8.799416720674012e-05,
|
| 92 |
+
"loss": 2.149,
|
| 93 |
+
"step": 3708
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.301312165883687,
|
| 97 |
+
"grad_norm": 0.03631984442472458,
|
| 98 |
+
"learning_rate": 8.699611147116008e-05,
|
| 99 |
+
"loss": 2.1605,
|
| 100 |
+
"step": 4017
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 1.4014255629353638,
|
| 104 |
+
"grad_norm": 0.03152529150247574,
|
| 105 |
+
"learning_rate": 8.59948152948801e-05,
|
| 106 |
+
"loss": 2.1436,
|
| 107 |
+
"step": 4326
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.5015389599870403,
|
| 111 |
+
"grad_norm": 0.05229083448648453,
|
| 112 |
+
"learning_rate": 8.499351911860013e-05,
|
| 113 |
+
"loss": 2.1794,
|
| 114 |
+
"step": 4635
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.601652357038717,
|
| 118 |
+
"grad_norm": 0.027535825967788696,
|
| 119 |
+
"learning_rate": 8.399222294232017e-05,
|
| 120 |
+
"loss": 2.1641,
|
| 121 |
+
"step": 4944
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 1.7017657540903937,
|
| 125 |
+
"grad_norm": 0.0046606422401964664,
|
| 126 |
+
"learning_rate": 8.299092676604019e-05,
|
| 127 |
+
"loss": 2.1242,
|
| 128 |
+
"step": 5253
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.8018791511420704,
|
| 132 |
+
"grad_norm": 0.040044769644737244,
|
| 133 |
+
"learning_rate": 8.198963058976021e-05,
|
| 134 |
+
"loss": 2.14,
|
| 135 |
+
"step": 5562
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 1.901992548193747,
|
| 139 |
+
"grad_norm": 0.01644105464220047,
|
| 140 |
+
"learning_rate": 8.098833441348024e-05,
|
| 141 |
+
"loss": 2.1488,
|
| 142 |
+
"step": 5871
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 2.001943949457314,
|
| 146 |
+
"grad_norm": 0.01979956403374672,
|
| 147 |
+
"learning_rate": 7.99902786779002e-05,
|
| 148 |
+
"loss": 2.1586,
|
| 149 |
+
"step": 6180
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 2.1020573465089907,
|
| 153 |
+
"grad_norm": 0.027056917548179626,
|
| 154 |
+
"learning_rate": 7.898898250162022e-05,
|
| 155 |
+
"loss": 2.1503,
|
| 156 |
+
"step": 6489
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 2.2021707435606674,
|
| 160 |
+
"grad_norm": 0.01640058308839798,
|
| 161 |
+
"learning_rate": 7.798768632534025e-05,
|
| 162 |
+
"loss": 2.1418,
|
| 163 |
+
"step": 6798
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 2.302284140612344,
|
| 167 |
+
"grad_norm": 0.01607314869761467,
|
| 168 |
+
"learning_rate": 7.698639014906027e-05,
|
| 169 |
+
"loss": 2.1514,
|
| 170 |
+
"step": 7107
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 2.4023975376640205,
|
| 174 |
+
"grad_norm": 0.017274878919124603,
|
| 175 |
+
"learning_rate": 7.59850939727803e-05,
|
| 176 |
+
"loss": 2.1465,
|
| 177 |
+
"step": 7416
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 2.5025109347156973,
|
| 181 |
+
"grad_norm": 0.024440627545118332,
|
| 182 |
+
"learning_rate": 7.498379779650033e-05,
|
| 183 |
+
"loss": 2.1211,
|
| 184 |
+
"step": 7725
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 2.602624331767374,
|
| 188 |
+
"grad_norm": 0.005127054639160633,
|
| 189 |
+
"learning_rate": 7.39857420609203e-05,
|
| 190 |
+
"loss": 2.1423,
|
| 191 |
+
"step": 8034
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 2.702737728819051,
|
| 195 |
+
"grad_norm": 0.05081469565629959,
|
| 196 |
+
"learning_rate": 7.298444588464032e-05,
|
| 197 |
+
"loss": 2.1705,
|
| 198 |
+
"step": 8343
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 2.8028511258707276,
|
| 202 |
+
"grad_norm": 0.01722005568444729,
|
| 203 |
+
"learning_rate": 7.198314970836034e-05,
|
| 204 |
+
"loss": 2.1462,
|
| 205 |
+
"step": 8652
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 2.902964522922404,
|
| 209 |
+
"grad_norm": 0.03728850930929184,
|
| 210 |
+
"learning_rate": 7.098185353208037e-05,
|
| 211 |
+
"loss": 2.1306,
|
| 212 |
+
"step": 8961
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 3.002915924185971,
|
| 216 |
+
"grad_norm": 0.016364697366952896,
|
| 217 |
+
"learning_rate": 6.998055735580039e-05,
|
| 218 |
+
"loss": 2.1639,
|
| 219 |
+
"step": 9270
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 3.103029321237648,
|
| 223 |
+
"grad_norm": 0.005412334576249123,
|
| 224 |
+
"learning_rate": 6.897926117952041e-05,
|
| 225 |
+
"loss": 2.1624,
|
| 226 |
+
"step": 9579
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 3.2031427182893246,
|
| 230 |
+
"grad_norm": 0.0071863215416669846,
|
| 231 |
+
"learning_rate": 6.797796500324045e-05,
|
| 232 |
+
"loss": 2.1682,
|
| 233 |
+
"step": 9888
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 3.303256115341001,
|
| 237 |
+
"grad_norm": 0.01707269623875618,
|
| 238 |
+
"learning_rate": 6.69799092676604e-05,
|
| 239 |
+
"loss": 2.129,
|
| 240 |
+
"step": 10197
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 3.4033695123926777,
|
| 244 |
+
"grad_norm": 0.0162820965051651,
|
| 245 |
+
"learning_rate": 6.597861309138042e-05,
|
| 246 |
+
"loss": 2.1552,
|
| 247 |
+
"step": 10506
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 3.5034829094443545,
|
| 251 |
+
"grad_norm": 0.005164678208529949,
|
| 252 |
+
"learning_rate": 6.497731691510046e-05,
|
| 253 |
+
"loss": 2.137,
|
| 254 |
+
"step": 10815
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 3.6035963064960312,
|
| 258 |
+
"grad_norm": 0.01631810888648033,
|
| 259 |
+
"learning_rate": 6.397602073882049e-05,
|
| 260 |
+
"loss": 2.1578,
|
| 261 |
+
"step": 11124
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 3.703709703547708,
|
| 265 |
+
"grad_norm": 0.004154821392148733,
|
| 266 |
+
"learning_rate": 6.297472456254051e-05,
|
| 267 |
+
"loss": 2.1602,
|
| 268 |
+
"step": 11433
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 3.8038231005993843,
|
| 272 |
+
"grad_norm": 0.004850070457905531,
|
| 273 |
+
"learning_rate": 6.197342838626053e-05,
|
| 274 |
+
"loss": 2.1412,
|
| 275 |
+
"step": 11742
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 3.903936497651061,
|
| 279 |
+
"grad_norm": 0.0028279961552470922,
|
| 280 |
+
"learning_rate": 6.0975372650680494e-05,
|
| 281 |
+
"loss": 2.1338,
|
| 282 |
+
"step": 12051
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 4.003887898914628,
|
| 286 |
+
"grad_norm": 0.016007574275135994,
|
| 287 |
+
"learning_rate": 5.997407647440052e-05,
|
| 288 |
+
"loss": 2.1753,
|
| 289 |
+
"step": 12360
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 4.104001295966305,
|
| 293 |
+
"grad_norm": 0.013178674504160881,
|
| 294 |
+
"learning_rate": 5.897278029812054e-05,
|
| 295 |
+
"loss": 2.1533,
|
| 296 |
+
"step": 12669
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 4.204114693017981,
|
| 300 |
+
"grad_norm": 0.010946434922516346,
|
| 301 |
+
"learning_rate": 5.797148412184057e-05,
|
| 302 |
+
"loss": 2.1506,
|
| 303 |
+
"step": 12978
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 4.304228090069659,
|
| 307 |
+
"grad_norm": 0.06032814085483551,
|
| 308 |
+
"learning_rate": 5.6970187945560595e-05,
|
| 309 |
+
"loss": 2.1408,
|
| 310 |
+
"step": 13287
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 4.404341487121335,
|
| 314 |
+
"grad_norm": 0.036340948194265366,
|
| 315 |
+
"learning_rate": 5.596889176928063e-05,
|
| 316 |
+
"loss": 2.1317,
|
| 317 |
+
"step": 13596
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 4.505750850477888,
|
| 321 |
+
"grad_norm": 0.022454094141721725,
|
| 322 |
+
"learning_rate": 5.4967595593000656e-05,
|
| 323 |
+
"loss": 2.1524,
|
| 324 |
+
"step": 13905
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 4.605864247529564,
|
| 328 |
+
"grad_norm": 0.01212249230593443,
|
| 329 |
+
"learning_rate": 5.396953985742061e-05,
|
| 330 |
+
"loss": 2.1681,
|
| 331 |
+
"step": 14214
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 4.705977644581241,
|
| 335 |
+
"grad_norm": 0.021747123450040817,
|
| 336 |
+
"learning_rate": 5.296824368114064e-05,
|
| 337 |
+
"loss": 2.1562,
|
| 338 |
+
"step": 14523
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 4.806091041632918,
|
| 342 |
+
"grad_norm": 0.005955239292234182,
|
| 343 |
+
"learning_rate": 5.196694750486067e-05,
|
| 344 |
+
"loss": 2.1828,
|
| 345 |
+
"step": 14832
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 4.906204438684594,
|
| 349 |
+
"grad_norm": 0.012121310457587242,
|
| 350 |
+
"learning_rate": 5.096565132858069e-05,
|
| 351 |
+
"loss": 2.1437,
|
| 352 |
+
"step": 15141
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 5.006479831524381,
|
| 356 |
+
"grad_norm": 0.015455316752195358,
|
| 357 |
+
"learning_rate": 4.996435515230072e-05,
|
| 358 |
+
"loss": 2.1528,
|
| 359 |
+
"step": 15450
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 5.106593228576057,
|
| 363 |
+
"grad_norm": 0.0007723022717982531,
|
| 364 |
+
"learning_rate": 4.896305897602074e-05,
|
| 365 |
+
"loss": 2.1723,
|
| 366 |
+
"step": 15759
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 5.206706625627733,
|
| 370 |
+
"grad_norm": 0.004882230423390865,
|
| 371 |
+
"learning_rate": 4.7965003240440704e-05,
|
| 372 |
+
"loss": 2.1525,
|
| 373 |
+
"step": 16068
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 5.30682002267941,
|
| 377 |
+
"grad_norm": 0.012511253356933594,
|
| 378 |
+
"learning_rate": 4.696370706416073e-05,
|
| 379 |
+
"loss": 2.1458,
|
| 380 |
+
"step": 16377
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 5.406933419731087,
|
| 384 |
+
"grad_norm": 0.0262750256806612,
|
| 385 |
+
"learning_rate": 4.596241088788075e-05,
|
| 386 |
+
"loss": 2.0936,
|
| 387 |
+
"step": 16686
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 5.507046816782764,
|
| 391 |
+
"grad_norm": 0.007834335789084435,
|
| 392 |
+
"learning_rate": 4.4961114711600785e-05,
|
| 393 |
+
"loss": 2.1322,
|
| 394 |
+
"step": 16995
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 5.60716021383444,
|
| 398 |
+
"grad_norm": 0.020112166181206703,
|
| 399 |
+
"learning_rate": 4.3959818535320806e-05,
|
| 400 |
+
"loss": 2.1223,
|
| 401 |
+
"step": 17304
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 5.707273610886117,
|
| 405 |
+
"grad_norm": 0.0009047465864568949,
|
| 406 |
+
"learning_rate": 4.295852235904083e-05,
|
| 407 |
+
"loss": 2.1425,
|
| 408 |
+
"step": 17613
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 5.807387007937794,
|
| 412 |
+
"grad_norm": 0.006675088778138161,
|
| 413 |
+
"learning_rate": 4.195722618276085e-05,
|
| 414 |
+
"loss": 2.1433,
|
| 415 |
+
"step": 17922
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 5.90750040498947,
|
| 419 |
+
"grad_norm": 0.022665197029709816,
|
| 420 |
+
"learning_rate": 4.0959170447180816e-05,
|
| 421 |
+
"loss": 2.1942,
|
| 422 |
+
"step": 18231
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 6.007451806253037,
|
| 426 |
+
"grad_norm": 0.039628468453884125,
|
| 427 |
+
"learning_rate": 3.995787427090085e-05,
|
| 428 |
+
"loss": 2.1444,
|
| 429 |
+
"step": 18540
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 6.107565203304714,
|
| 433 |
+
"grad_norm": 0.0011520631378516555,
|
| 434 |
+
"learning_rate": 3.895657809462087e-05,
|
| 435 |
+
"loss": 2.1524,
|
| 436 |
+
"step": 18849
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 6.207678600356391,
|
| 440 |
+
"grad_norm": 0.03555034101009369,
|
| 441 |
+
"learning_rate": 3.79552819183409e-05,
|
| 442 |
+
"loss": 2.1203,
|
| 443 |
+
"step": 19158
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 6.307791997408067,
|
| 447 |
+
"grad_norm": 0.013503223657608032,
|
| 448 |
+
"learning_rate": 3.695398574206092e-05,
|
| 449 |
+
"loss": 2.1352,
|
| 450 |
+
"step": 19467
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 6.407905394459744,
|
| 454 |
+
"grad_norm": 0.026100030168890953,
|
| 455 |
+
"learning_rate": 3.595268956578095e-05,
|
| 456 |
+
"loss": 2.1655,
|
| 457 |
+
"step": 19776
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 6.508018791511421,
|
| 461 |
+
"grad_norm": 0.0006254952168092132,
|
| 462 |
+
"learning_rate": 3.4954633830200915e-05,
|
| 463 |
+
"loss": 2.1352,
|
| 464 |
+
"step": 20085
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 6.608132188563097,
|
| 468 |
+
"grad_norm": 0.01487251278012991,
|
| 469 |
+
"learning_rate": 3.3953337653920935e-05,
|
| 470 |
+
"loss": 2.1799,
|
| 471 |
+
"step": 20394
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 6.708245585614774,
|
| 475 |
+
"grad_norm": 0.0053134192712605,
|
| 476 |
+
"learning_rate": 3.295204147764096e-05,
|
| 477 |
+
"loss": 2.1506,
|
| 478 |
+
"step": 20703
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 6.808358982666451,
|
| 482 |
+
"grad_norm": 0.006517978850752115,
|
| 483 |
+
"learning_rate": 3.195074530136098e-05,
|
| 484 |
+
"loss": 2.1567,
|
| 485 |
+
"step": 21012
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 6.908472379718127,
|
| 489 |
+
"grad_norm": 0.014856001362204552,
|
| 490 |
+
"learning_rate": 3.0949449125081016e-05,
|
| 491 |
+
"loss": 2.1593,
|
| 492 |
+
"step": 21321
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 7.008423780981694,
|
| 496 |
+
"grad_norm": 0.0019718091934919357,
|
| 497 |
+
"learning_rate": 2.994815294880104e-05,
|
| 498 |
+
"loss": 2.1634,
|
| 499 |
+
"step": 21630
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 7.108537178033371,
|
| 503 |
+
"grad_norm": 0.017417173832654953,
|
| 504 |
+
"learning_rate": 2.8946856772521063e-05,
|
| 505 |
+
"loss": 2.1433,
|
| 506 |
+
"step": 21939
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 7.208650575085048,
|
| 510 |
+
"grad_norm": 0.034113720059394836,
|
| 511 |
+
"learning_rate": 2.7948801036941023e-05,
|
| 512 |
+
"loss": 2.1489,
|
| 513 |
+
"step": 22248
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 7.308763972136725,
|
| 517 |
+
"grad_norm": 0.018758224323391914,
|
| 518 |
+
"learning_rate": 2.694750486066105e-05,
|
| 519 |
+
"loss": 2.13,
|
| 520 |
+
"step": 22557
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 7.408877369188401,
|
| 524 |
+
"grad_norm": 0.04419185221195221,
|
| 525 |
+
"learning_rate": 2.594620868438108e-05,
|
| 526 |
+
"loss": 2.1641,
|
| 527 |
+
"step": 22866
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 7.508990766240077,
|
| 531 |
+
"grad_norm": 0.001091700978577137,
|
| 532 |
+
"learning_rate": 2.4944912508101104e-05,
|
| 533 |
+
"loss": 2.1763,
|
| 534 |
+
"step": 23175
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 7.609104163291755,
|
| 538 |
+
"grad_norm": 0.0011129506165161729,
|
| 539 |
+
"learning_rate": 2.3943616331821128e-05,
|
| 540 |
+
"loss": 2.125,
|
| 541 |
+
"step": 23484
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 7.709217560343431,
|
| 545 |
+
"grad_norm": 0.007060033269226551,
|
| 546 |
+
"learning_rate": 2.2942320155541155e-05,
|
| 547 |
+
"loss": 2.1357,
|
| 548 |
+
"step": 23793
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 7.809330957395108,
|
| 552 |
+
"grad_norm": 0.004454698413610458,
|
| 553 |
+
"learning_rate": 2.1944264419961118e-05,
|
| 554 |
+
"loss": 2.1502,
|
| 555 |
+
"step": 24102
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 7.9094443544467845,
|
| 559 |
+
"grad_norm": 0.014023036696016788,
|
| 560 |
+
"learning_rate": 2.0942968243681142e-05,
|
| 561 |
+
"loss": 2.1397,
|
| 562 |
+
"step": 24411
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 8.00939575571035,
|
| 566 |
+
"grad_norm": 0.0020535311195999384,
|
| 567 |
+
"learning_rate": 1.994167206740117e-05,
|
| 568 |
+
"loss": 2.1594,
|
| 569 |
+
"step": 24720
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 8.109509152762028,
|
| 573 |
+
"grad_norm": 0.030191343277692795,
|
| 574 |
+
"learning_rate": 1.8940375891121192e-05,
|
| 575 |
+
"loss": 2.1379,
|
| 576 |
+
"step": 25029
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 8.209622549813705,
|
| 580 |
+
"grad_norm": 0.02658534049987793,
|
| 581 |
+
"learning_rate": 1.793907971484122e-05,
|
| 582 |
+
"loss": 2.163,
|
| 583 |
+
"step": 25338
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 8.30973594686538,
|
| 587 |
+
"grad_norm": 0.006291504483669996,
|
| 588 |
+
"learning_rate": 1.6937783538561243e-05,
|
| 589 |
+
"loss": 2.1445,
|
| 590 |
+
"step": 25647
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 8.409849343917058,
|
| 594 |
+
"grad_norm": 0.013936794362962246,
|
| 595 |
+
"learning_rate": 1.593648736228127e-05,
|
| 596 |
+
"loss": 2.1571,
|
| 597 |
+
"step": 25956
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 8.509962740968735,
|
| 601 |
+
"grad_norm": 0.031892433762550354,
|
| 602 |
+
"learning_rate": 1.4938431626701233e-05,
|
| 603 |
+
"loss": 2.1562,
|
| 604 |
+
"step": 26265
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 8.610076138020412,
|
| 608 |
+
"grad_norm": 0.0028675836510956287,
|
| 609 |
+
"learning_rate": 1.3937135450421257e-05,
|
| 610 |
+
"loss": 2.1373,
|
| 611 |
+
"step": 26574
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 8.710189535072088,
|
| 615 |
+
"grad_norm": 0.027382852509617805,
|
| 616 |
+
"learning_rate": 1.2935839274141284e-05,
|
| 617 |
+
"loss": 2.1162,
|
| 618 |
+
"step": 26883
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 8.810302932123765,
|
| 622 |
+
"grad_norm": 0.006504488177597523,
|
| 623 |
+
"learning_rate": 1.193454309786131e-05,
|
| 624 |
+
"loss": 2.1281,
|
| 625 |
+
"step": 27192
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 8.910416329175442,
|
| 629 |
+
"grad_norm": 0.006998216733336449,
|
| 630 |
+
"learning_rate": 1.0933246921581337e-05,
|
| 631 |
+
"loss": 2.1681,
|
| 632 |
+
"step": 27501
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 9.01036773043901,
|
| 636 |
+
"grad_norm": 0.0019873257260769606,
|
| 637 |
+
"learning_rate": 9.931950745301362e-06,
|
| 638 |
+
"loss": 2.1397,
|
| 639 |
+
"step": 27810
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 9.110481127490685,
|
| 643 |
+
"grad_norm": 0.04183882847428322,
|
| 644 |
+
"learning_rate": 8.933895009721323e-06,
|
| 645 |
+
"loss": 2.1425,
|
| 646 |
+
"step": 28119
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 9.210594524542362,
|
| 650 |
+
"grad_norm": 0.008224571123719215,
|
| 651 |
+
"learning_rate": 7.932598833441349e-06,
|
| 652 |
+
"loss": 2.1567,
|
| 653 |
+
"step": 28428
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 9.31070792159404,
|
| 657 |
+
"grad_norm": 0.009601627476513386,
|
| 658 |
+
"learning_rate": 6.931302657161375e-06,
|
| 659 |
+
"loss": 2.1334,
|
| 660 |
+
"step": 28737
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 9.410821318645715,
|
| 664 |
+
"grad_norm": 0.028333676978945732,
|
| 665 |
+
"learning_rate": 5.9300064808814e-06,
|
| 666 |
+
"loss": 2.1732,
|
| 667 |
+
"step": 29046
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 9.510934715697392,
|
| 671 |
+
"grad_norm": 0.01850961521267891,
|
| 672 |
+
"learning_rate": 4.9287103046014265e-06,
|
| 673 |
+
"loss": 2.1401,
|
| 674 |
+
"step": 29355
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 9.611048112749069,
|
| 678 |
+
"grad_norm": 0.0022975043393671513,
|
| 679 |
+
"learning_rate": 3.927414128321452e-06,
|
| 680 |
+
"loss": 2.1523,
|
| 681 |
+
"step": 29664
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 9.711161509800744,
|
| 685 |
+
"grad_norm": 0.033216096460819244,
|
| 686 |
+
"learning_rate": 2.9261179520414777e-06,
|
| 687 |
+
"loss": 2.1497,
|
| 688 |
+
"step": 29973
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 9.811274906852422,
|
| 692 |
+
"grad_norm": 0.024143142625689507,
|
| 693 |
+
"learning_rate": 1.928062216461439e-06,
|
| 694 |
+
"loss": 2.1504,
|
| 695 |
+
"step": 30282
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 9.911388303904099,
|
| 699 |
+
"grad_norm": 0.0085253044962883,
|
| 700 |
+
"learning_rate": 9.267660401814646e-07,
|
| 701 |
+
"loss": 2.177,
|
| 702 |
+
"step": 30591
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 9.998542037907015,
|
| 706 |
+
"step": 30860,
|
| 707 |
+
"total_flos": 1.461291118888168e+18,
|
| 708 |
+
"train_loss": 1.202193304075487,
|
| 709 |
+
"train_runtime": 14493.7066,
|
| 710 |
+
"train_samples_per_second": 255.536,
|
| 711 |
+
"train_steps_per_second": 2.129
|
| 712 |
+
}
|
| 713 |
+
],
|
| 714 |
+
"logging_steps": 309,
|
| 715 |
+
"max_steps": 30860,
|
| 716 |
+
"num_input_tokens_seen": 0,
|
| 717 |
+
"num_train_epochs": 10,
|
| 718 |
+
"save_steps": 6800,
|
| 719 |
+
"stateful_callbacks": {
|
| 720 |
+
"TrainerControl": {
|
| 721 |
+
"args": {
|
| 722 |
+
"should_epoch_stop": false,
|
| 723 |
+
"should_evaluate": false,
|
| 724 |
+
"should_log": false,
|
| 725 |
+
"should_save": true,
|
| 726 |
+
"should_training_stop": true
|
| 727 |
+
},
|
| 728 |
+
"attributes": {}
|
| 729 |
+
}
|
| 730 |
+
},
|
| 731 |
+
"total_flos": 1.461291118888168e+18,
|
| 732 |
+
"train_batch_size": 30,
|
| 733 |
+
"trial_name": null,
|
| 734 |
+
"trial_params": null
|
| 735 |
+
}
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|