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 +743 -0
- vocab.txt +0 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 99.91201764057332,
|
| 3 |
+
"total_flos": 5.364383961180322e+18,
|
| 4 |
+
"train_loss": 1.6900030501417007,
|
| 5 |
+
"train_runtime": 96835.149,
|
| 6 |
+
"train_samples_per_second": 140.49,
|
| 7 |
+
"train_steps_per_second": 1.17
|
| 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": 99.91201764057332,
|
| 3 |
+
"total_flos": 5.364383961180322e+18,
|
| 4 |
+
"train_loss": 1.6900030501417007,
|
| 5 |
+
"train_runtime": 96835.149,
|
| 6 |
+
"train_samples_per_second": 140.49,
|
| 7 |
+
"train_steps_per_second": 1.17
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,743 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 99.91201764057332,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 113300,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.9993384785005512,
|
| 14 |
+
"grad_norm": 6.264042854309082,
|
| 15 |
+
"learning_rate": 9.90035304501324e-06,
|
| 16 |
+
"loss": 2.1264,
|
| 17 |
+
"step": 1133
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 1.9984564498346198,
|
| 21 |
+
"grad_norm": 8.962105751037598,
|
| 22 |
+
"learning_rate": 9.80035304501324e-06,
|
| 23 |
+
"loss": 2.0771,
|
| 24 |
+
"step": 2266
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 2.997574421168688,
|
| 28 |
+
"grad_norm": 13.9079008102417,
|
| 29 |
+
"learning_rate": 9.70044130626655e-06,
|
| 30 |
+
"loss": 2.0072,
|
| 31 |
+
"step": 3399
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 3.9966923925027564,
|
| 35 |
+
"grad_norm": 9.212200164794922,
|
| 36 |
+
"learning_rate": 9.60044130626655e-06,
|
| 37 |
+
"loss": 1.9554,
|
| 38 |
+
"step": 4532
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 4.995810363836824,
|
| 42 |
+
"grad_norm": 9.967486381530762,
|
| 43 |
+
"learning_rate": 9.50052956751986e-06,
|
| 44 |
+
"loss": 1.9156,
|
| 45 |
+
"step": 5665
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 5.9949283351708935,
|
| 49 |
+
"grad_norm": 9.588685989379883,
|
| 50 |
+
"learning_rate": 9.40061782877317e-06,
|
| 51 |
+
"loss": 1.9031,
|
| 52 |
+
"step": 6798
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 6.994046306504961,
|
| 56 |
+
"grad_norm": 5.2934889793396,
|
| 57 |
+
"learning_rate": 9.30061782877317e-06,
|
| 58 |
+
"loss": 1.8739,
|
| 59 |
+
"step": 7931
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 7.99316427783903,
|
| 63 |
+
"grad_norm": 12.791667938232422,
|
| 64 |
+
"learning_rate": 9.20070609002648e-06,
|
| 65 |
+
"loss": 1.8476,
|
| 66 |
+
"step": 9064
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 8.992282249173098,
|
| 70 |
+
"grad_norm": 5.592958927154541,
|
| 71 |
+
"learning_rate": 9.100706090026479e-06,
|
| 72 |
+
"loss": 1.8378,
|
| 73 |
+
"step": 10197
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 9.991400220507167,
|
| 77 |
+
"grad_norm": 6.236687660217285,
|
| 78 |
+
"learning_rate": 9.000706090026478e-06,
|
| 79 |
+
"loss": 1.8141,
|
| 80 |
+
"step": 11330
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 10.990518191841234,
|
| 84 |
+
"grad_norm": 14.459975242614746,
|
| 85 |
+
"learning_rate": 8.90079435127979e-06,
|
| 86 |
+
"loss": 1.8008,
|
| 87 |
+
"step": 12463
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 11.989636163175303,
|
| 91 |
+
"grad_norm": 7.29625940322876,
|
| 92 |
+
"learning_rate": 8.800794351279789e-06,
|
| 93 |
+
"loss": 1.7965,
|
| 94 |
+
"step": 13596
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 12.98875413450937,
|
| 98 |
+
"grad_norm": 7.927811622619629,
|
| 99 |
+
"learning_rate": 8.7008826125331e-06,
|
| 100 |
+
"loss": 1.7768,
|
| 101 |
+
"step": 14729
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 13.98787210584344,
|
| 105 |
+
"grad_norm": 7.546788692474365,
|
| 106 |
+
"learning_rate": 8.600882612533099e-06,
|
| 107 |
+
"loss": 1.7735,
|
| 108 |
+
"step": 15862
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 14.986990077177508,
|
| 112 |
+
"grad_norm": 2.396638870239258,
|
| 113 |
+
"learning_rate": 8.500970873786408e-06,
|
| 114 |
+
"loss": 1.7688,
|
| 115 |
+
"step": 16995
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 15.986108048511577,
|
| 119 |
+
"grad_norm": 4.344698905944824,
|
| 120 |
+
"learning_rate": 8.400970873786408e-06,
|
| 121 |
+
"loss": 1.7521,
|
| 122 |
+
"step": 18128
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 16.985226019845644,
|
| 126 |
+
"grad_norm": 3.3172178268432617,
|
| 127 |
+
"learning_rate": 8.300970873786409e-06,
|
| 128 |
+
"loss": 1.7524,
|
| 129 |
+
"step": 19261
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 17.984343991179713,
|
| 133 |
+
"grad_norm": 4.233256816864014,
|
| 134 |
+
"learning_rate": 8.200970873786408e-06,
|
| 135 |
+
"loss": 1.7394,
|
| 136 |
+
"step": 20394
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 18.983461962513783,
|
| 140 |
+
"grad_norm": 3.924440860748291,
|
| 141 |
+
"learning_rate": 8.101147396293027e-06,
|
| 142 |
+
"loss": 1.7355,
|
| 143 |
+
"step": 21527
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 19.982579933847852,
|
| 147 |
+
"grad_norm": 8.452503204345703,
|
| 148 |
+
"learning_rate": 8.001147396293027e-06,
|
| 149 |
+
"loss": 1.7245,
|
| 150 |
+
"step": 22660
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 20.981697905181917,
|
| 154 |
+
"grad_norm": 10.714750289916992,
|
| 155 |
+
"learning_rate": 7.901147396293028e-06,
|
| 156 |
+
"loss": 1.7205,
|
| 157 |
+
"step": 23793
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 21.980815876515987,
|
| 161 |
+
"grad_norm": 12.158623695373535,
|
| 162 |
+
"learning_rate": 7.801147396293029e-06,
|
| 163 |
+
"loss": 1.7207,
|
| 164 |
+
"step": 24926
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 22.979933847850056,
|
| 168 |
+
"grad_norm": 2.1460916996002197,
|
| 169 |
+
"learning_rate": 7.701323918799648e-06,
|
| 170 |
+
"loss": 1.716,
|
| 171 |
+
"step": 26059
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 23.979051819184125,
|
| 175 |
+
"grad_norm": 2.6795060634613037,
|
| 176 |
+
"learning_rate": 7.601323918799648e-06,
|
| 177 |
+
"loss": 1.71,
|
| 178 |
+
"step": 27192
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 24.97816979051819,
|
| 182 |
+
"grad_norm": 4.405874729156494,
|
| 183 |
+
"learning_rate": 7.501412180052957e-06,
|
| 184 |
+
"loss": 1.7143,
|
| 185 |
+
"step": 28325
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 25.97728776185226,
|
| 189 |
+
"grad_norm": 2.3188281059265137,
|
| 190 |
+
"learning_rate": 7.401412180052957e-06,
|
| 191 |
+
"loss": 1.6879,
|
| 192 |
+
"step": 29458
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 26.97640573318633,
|
| 196 |
+
"grad_norm": 4.247509002685547,
|
| 197 |
+
"learning_rate": 7.301412180052957e-06,
|
| 198 |
+
"loss": 1.7008,
|
| 199 |
+
"step": 30591
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 27.9755237045204,
|
| 203 |
+
"grad_norm": 5.176025390625,
|
| 204 |
+
"learning_rate": 7.2015004413062674e-06,
|
| 205 |
+
"loss": 1.6904,
|
| 206 |
+
"step": 31724
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 28.974641675854464,
|
| 210 |
+
"grad_norm": 3.4096691608428955,
|
| 211 |
+
"learning_rate": 7.1015004413062675e-06,
|
| 212 |
+
"loss": 1.6884,
|
| 213 |
+
"step": 32857
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 29.973759647188533,
|
| 217 |
+
"grad_norm": 5.346144676208496,
|
| 218 |
+
"learning_rate": 7.001588702559578e-06,
|
| 219 |
+
"loss": 1.6874,
|
| 220 |
+
"step": 33990
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 30.972877618522602,
|
| 224 |
+
"grad_norm": 4.809396743774414,
|
| 225 |
+
"learning_rate": 6.901588702559577e-06,
|
| 226 |
+
"loss": 1.6834,
|
| 227 |
+
"step": 35123
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 31.97199558985667,
|
| 231 |
+
"grad_norm": 4.950044631958008,
|
| 232 |
+
"learning_rate": 6.8016769638128865e-06,
|
| 233 |
+
"loss": 1.6704,
|
| 234 |
+
"step": 36256
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 32.97111356119074,
|
| 238 |
+
"grad_norm": 2.259593963623047,
|
| 239 |
+
"learning_rate": 6.701676963812887e-06,
|
| 240 |
+
"loss": 1.676,
|
| 241 |
+
"step": 37389
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 33.97023153252481,
|
| 245 |
+
"grad_norm": 1.9857131242752075,
|
| 246 |
+
"learning_rate": 6.601765225066197e-06,
|
| 247 |
+
"loss": 1.6721,
|
| 248 |
+
"step": 38522
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 34.96934950385887,
|
| 252 |
+
"grad_norm": 2.6523988246917725,
|
| 253 |
+
"learning_rate": 6.501765225066197e-06,
|
| 254 |
+
"loss": 1.6519,
|
| 255 |
+
"step": 39655
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 35.968467475192945,
|
| 259 |
+
"grad_norm": 2.5178589820861816,
|
| 260 |
+
"learning_rate": 6.401853486319506e-06,
|
| 261 |
+
"loss": 1.6763,
|
| 262 |
+
"step": 40788
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 36.96758544652701,
|
| 266 |
+
"grad_norm": 3.966003894805908,
|
| 267 |
+
"learning_rate": 6.301941747572816e-06,
|
| 268 |
+
"loss": 1.6729,
|
| 269 |
+
"step": 41921
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 37.96670341786108,
|
| 273 |
+
"grad_norm": 2.874756336212158,
|
| 274 |
+
"learning_rate": 6.201941747572816e-06,
|
| 275 |
+
"loss": 1.6761,
|
| 276 |
+
"step": 43054
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 38.96582138919515,
|
| 280 |
+
"grad_norm": 5.206564426422119,
|
| 281 |
+
"learning_rate": 6.102030008826125e-06,
|
| 282 |
+
"loss": 1.6731,
|
| 283 |
+
"step": 44187
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 39.964939360529215,
|
| 287 |
+
"grad_norm": 3.3622446060180664,
|
| 288 |
+
"learning_rate": 6.002030008826126e-06,
|
| 289 |
+
"loss": 1.6649,
|
| 290 |
+
"step": 45320
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 40.96405733186329,
|
| 294 |
+
"grad_norm": 2.243980884552002,
|
| 295 |
+
"learning_rate": 5.902118270079436e-06,
|
| 296 |
+
"loss": 1.6617,
|
| 297 |
+
"step": 46453
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 41.96317530319735,
|
| 301 |
+
"grad_norm": 2.656219720840454,
|
| 302 |
+
"learning_rate": 5.802118270079435e-06,
|
| 303 |
+
"loss": 1.6539,
|
| 304 |
+
"step": 47586
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 42.96229327453142,
|
| 308 |
+
"grad_norm": 2.1969571113586426,
|
| 309 |
+
"learning_rate": 5.702206531332745e-06,
|
| 310 |
+
"loss": 1.6507,
|
| 311 |
+
"step": 48719
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 43.96141124586549,
|
| 315 |
+
"grad_norm": 2.570960760116577,
|
| 316 |
+
"learning_rate": 5.602294792586055e-06,
|
| 317 |
+
"loss": 1.6517,
|
| 318 |
+
"step": 49852
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 44.96052921719956,
|
| 322 |
+
"grad_norm": 2.2769925594329834,
|
| 323 |
+
"learning_rate": 5.502294792586055e-06,
|
| 324 |
+
"loss": 1.6622,
|
| 325 |
+
"step": 50985
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 45.95964718853363,
|
| 329 |
+
"grad_norm": 1.3121511936187744,
|
| 330 |
+
"learning_rate": 5.402383053839365e-06,
|
| 331 |
+
"loss": 1.6583,
|
| 332 |
+
"step": 52118
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 46.958765159867696,
|
| 336 |
+
"grad_norm": 2.186378002166748,
|
| 337 |
+
"learning_rate": 5.302383053839365e-06,
|
| 338 |
+
"loss": 1.6579,
|
| 339 |
+
"step": 53251
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 47.95788313120176,
|
| 343 |
+
"grad_norm": 1.257639765739441,
|
| 344 |
+
"learning_rate": 5.202471315092674e-06,
|
| 345 |
+
"loss": 1.653,
|
| 346 |
+
"step": 54384
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 48.957001102535834,
|
| 350 |
+
"grad_norm": 2.570171594619751,
|
| 351 |
+
"learning_rate": 5.102471315092676e-06,
|
| 352 |
+
"loss": 1.6503,
|
| 353 |
+
"step": 55517
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 49.9561190738699,
|
| 357 |
+
"grad_norm": 1.8046927452087402,
|
| 358 |
+
"learning_rate": 5.002559576345984e-06,
|
| 359 |
+
"loss": 1.6569,
|
| 360 |
+
"step": 56650
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 50.95523704520397,
|
| 364 |
+
"grad_norm": 2.055699348449707,
|
| 365 |
+
"learning_rate": 4.9026478375992945e-06,
|
| 366 |
+
"loss": 1.649,
|
| 367 |
+
"step": 57783
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 51.95435501653804,
|
| 371 |
+
"grad_norm": 1.5541220903396606,
|
| 372 |
+
"learning_rate": 4.802647837599294e-06,
|
| 373 |
+
"loss": 1.6463,
|
| 374 |
+
"step": 58916
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 52.953472987872104,
|
| 378 |
+
"grad_norm": 1.3364524841308594,
|
| 379 |
+
"learning_rate": 4.702736098852604e-06,
|
| 380 |
+
"loss": 1.6406,
|
| 381 |
+
"step": 60049
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 53.95259095920618,
|
| 385 |
+
"grad_norm": 1.8428481817245483,
|
| 386 |
+
"learning_rate": 4.602736098852604e-06,
|
| 387 |
+
"loss": 1.6443,
|
| 388 |
+
"step": 61182
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 54.95170893054024,
|
| 392 |
+
"grad_norm": 1.3728575706481934,
|
| 393 |
+
"learning_rate": 4.5028243601059144e-06,
|
| 394 |
+
"loss": 1.6427,
|
| 395 |
+
"step": 62315
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 55.95082690187431,
|
| 399 |
+
"grad_norm": 2.039762496948242,
|
| 400 |
+
"learning_rate": 4.402824360105914e-06,
|
| 401 |
+
"loss": 1.6523,
|
| 402 |
+
"step": 63448
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 56.94994487320838,
|
| 406 |
+
"grad_norm": 0.4440517723560333,
|
| 407 |
+
"learning_rate": 4.302912621359224e-06,
|
| 408 |
+
"loss": 1.6346,
|
| 409 |
+
"step": 64581
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 57.949062844542446,
|
| 413 |
+
"grad_norm": 2.8719191551208496,
|
| 414 |
+
"learning_rate": 4.202912621359223e-06,
|
| 415 |
+
"loss": 1.6481,
|
| 416 |
+
"step": 65714
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 58.94818081587652,
|
| 420 |
+
"grad_norm": 2.116090774536133,
|
| 421 |
+
"learning_rate": 4.1030008826125335e-06,
|
| 422 |
+
"loss": 1.6403,
|
| 423 |
+
"step": 66847
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 59.947298787210585,
|
| 427 |
+
"grad_norm": 2.5801730155944824,
|
| 428 |
+
"learning_rate": 4.003089143865843e-06,
|
| 429 |
+
"loss": 1.6445,
|
| 430 |
+
"step": 67980
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 60.94641675854465,
|
| 434 |
+
"grad_norm": 1.5309211015701294,
|
| 435 |
+
"learning_rate": 3.903089143865843e-06,
|
| 436 |
+
"loss": 1.6495,
|
| 437 |
+
"step": 69113
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 61.94553472987872,
|
| 441 |
+
"grad_norm": 2.1777145862579346,
|
| 442 |
+
"learning_rate": 3.803177405119153e-06,
|
| 443 |
+
"loss": 1.6477,
|
| 444 |
+
"step": 70246
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 62.94465270121279,
|
| 448 |
+
"grad_norm": 1.003503441810608,
|
| 449 |
+
"learning_rate": 3.7031774051191535e-06,
|
| 450 |
+
"loss": 1.6387,
|
| 451 |
+
"step": 71379
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 63.943770672546854,
|
| 455 |
+
"grad_norm": 2.015745162963867,
|
| 456 |
+
"learning_rate": 3.603265666372463e-06,
|
| 457 |
+
"loss": 1.6307,
|
| 458 |
+
"step": 72512
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 64.94288864388092,
|
| 462 |
+
"grad_norm": 2.482320547103882,
|
| 463 |
+
"learning_rate": 3.503265666372463e-06,
|
| 464 |
+
"loss": 1.6402,
|
| 465 |
+
"step": 73645
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 65.94200661521499,
|
| 469 |
+
"grad_norm": 2.404247999191284,
|
| 470 |
+
"learning_rate": 3.403353927625773e-06,
|
| 471 |
+
"loss": 1.6409,
|
| 472 |
+
"step": 74778
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 66.94112458654907,
|
| 476 |
+
"grad_norm": 1.197210669517517,
|
| 477 |
+
"learning_rate": 3.3034421888790823e-06,
|
| 478 |
+
"loss": 1.6518,
|
| 479 |
+
"step": 75911
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 67.94024255788312,
|
| 483 |
+
"grad_norm": 0.6208453178405762,
|
| 484 |
+
"learning_rate": 3.2034421888790824e-06,
|
| 485 |
+
"loss": 1.6461,
|
| 486 |
+
"step": 77044
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 68.9393605292172,
|
| 490 |
+
"grad_norm": 1.967084527015686,
|
| 491 |
+
"learning_rate": 3.1035304501323922e-06,
|
| 492 |
+
"loss": 1.6393,
|
| 493 |
+
"step": 78177
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 69.93847850055127,
|
| 497 |
+
"grad_norm": 1.0003052949905396,
|
| 498 |
+
"learning_rate": 3.003530450132392e-06,
|
| 499 |
+
"loss": 1.6259,
|
| 500 |
+
"step": 79310
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 70.93759647188534,
|
| 504 |
+
"grad_norm": 0.6473856568336487,
|
| 505 |
+
"learning_rate": 2.9036187113857018e-06,
|
| 506 |
+
"loss": 1.6302,
|
| 507 |
+
"step": 80443
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 71.9367144432194,
|
| 511 |
+
"grad_norm": 0.949741780757904,
|
| 512 |
+
"learning_rate": 2.8036187113857015e-06,
|
| 513 |
+
"loss": 1.6355,
|
| 514 |
+
"step": 81576
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 72.93583241455347,
|
| 518 |
+
"grad_norm": 1.2901132106781006,
|
| 519 |
+
"learning_rate": 2.7037069726390117e-06,
|
| 520 |
+
"loss": 1.6335,
|
| 521 |
+
"step": 82709
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 73.93495038588755,
|
| 525 |
+
"grad_norm": 1.8243396282196045,
|
| 526 |
+
"learning_rate": 2.6037952338923216e-06,
|
| 527 |
+
"loss": 1.6329,
|
| 528 |
+
"step": 83842
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 74.9340683572216,
|
| 532 |
+
"grad_norm": 0.6137579679489136,
|
| 533 |
+
"learning_rate": 2.5037952338923217e-06,
|
| 534 |
+
"loss": 1.6421,
|
| 535 |
+
"step": 84975
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 75.93318632855568,
|
| 539 |
+
"grad_norm": 1.270875096321106,
|
| 540 |
+
"learning_rate": 2.403883495145631e-06,
|
| 541 |
+
"loss": 1.628,
|
| 542 |
+
"step": 86108
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 76.93230429988975,
|
| 546 |
+
"grad_norm": 0.9131399393081665,
|
| 547 |
+
"learning_rate": 2.3038834951456313e-06,
|
| 548 |
+
"loss": 1.6235,
|
| 549 |
+
"step": 87241
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 77.93142227122381,
|
| 553 |
+
"grad_norm": 1.042668342590332,
|
| 554 |
+
"learning_rate": 2.203971756398941e-06,
|
| 555 |
+
"loss": 1.6323,
|
| 556 |
+
"step": 88374
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 78.93054024255788,
|
| 560 |
+
"grad_norm": 2.174466609954834,
|
| 561 |
+
"learning_rate": 2.103971756398941e-06,
|
| 562 |
+
"loss": 1.6263,
|
| 563 |
+
"step": 89507
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 79.92965821389195,
|
| 567 |
+
"grad_norm": 0.5852002501487732,
|
| 568 |
+
"learning_rate": 2.004060017652251e-06,
|
| 569 |
+
"loss": 1.6354,
|
| 570 |
+
"step": 90640
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 80.92877618522601,
|
| 574 |
+
"grad_norm": 0.41708850860595703,
|
| 575 |
+
"learning_rate": 1.9040600176522508e-06,
|
| 576 |
+
"loss": 1.6177,
|
| 577 |
+
"step": 91773
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 81.92789415656009,
|
| 581 |
+
"grad_norm": 1.2882851362228394,
|
| 582 |
+
"learning_rate": 1.8041482789055606e-06,
|
| 583 |
+
"loss": 1.6317,
|
| 584 |
+
"step": 92906
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 82.92701212789416,
|
| 588 |
+
"grad_norm": 1.3567571640014648,
|
| 589 |
+
"learning_rate": 1.7042365401588703e-06,
|
| 590 |
+
"loss": 1.6336,
|
| 591 |
+
"step": 94039
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 83.92613009922823,
|
| 595 |
+
"grad_norm": 1.7891902923583984,
|
| 596 |
+
"learning_rate": 1.6042365401588702e-06,
|
| 597 |
+
"loss": 1.6285,
|
| 598 |
+
"step": 95172
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 84.92524807056229,
|
| 602 |
+
"grad_norm": 1.0167814493179321,
|
| 603 |
+
"learning_rate": 1.5043248014121802e-06,
|
| 604 |
+
"loss": 1.6233,
|
| 605 |
+
"step": 96305
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 85.92436604189636,
|
| 609 |
+
"grad_norm": 0.6127368211746216,
|
| 610 |
+
"learning_rate": 1.4043248014121801e-06,
|
| 611 |
+
"loss": 1.6194,
|
| 612 |
+
"step": 97438
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 86.92348401323044,
|
| 616 |
+
"grad_norm": 0.4271755516529083,
|
| 617 |
+
"learning_rate": 1.30441306266549e-06,
|
| 618 |
+
"loss": 1.6292,
|
| 619 |
+
"step": 98571
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 87.9226019845645,
|
| 623 |
+
"grad_norm": 0.6342440843582153,
|
| 624 |
+
"learning_rate": 1.20441306266549e-06,
|
| 625 |
+
"loss": 1.6266,
|
| 626 |
+
"step": 99704
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 88.92171995589857,
|
| 630 |
+
"grad_norm": 0.4804386794567108,
|
| 631 |
+
"learning_rate": 1.1045013239187997e-06,
|
| 632 |
+
"loss": 1.624,
|
| 633 |
+
"step": 100837
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 89.92083792723264,
|
| 637 |
+
"grad_norm": 0.6205073595046997,
|
| 638 |
+
"learning_rate": 1.0045895851721096e-06,
|
| 639 |
+
"loss": 1.6225,
|
| 640 |
+
"step": 101970
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 90.9199558985667,
|
| 644 |
+
"grad_norm": 0.5994776487350464,
|
| 645 |
+
"learning_rate": 9.045895851721096e-07,
|
| 646 |
+
"loss": 1.6296,
|
| 647 |
+
"step": 103103
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 91.91907386990077,
|
| 651 |
+
"grad_norm": 1.454080581665039,
|
| 652 |
+
"learning_rate": 8.046778464254193e-07,
|
| 653 |
+
"loss": 1.6254,
|
| 654 |
+
"step": 104236
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 92.91819184123484,
|
| 658 |
+
"grad_norm": 0.8811701536178589,
|
| 659 |
+
"learning_rate": 7.046778464254193e-07,
|
| 660 |
+
"loss": 1.6295,
|
| 661 |
+
"step": 105369
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 93.9173098125689,
|
| 665 |
+
"grad_norm": 0.6332120895385742,
|
| 666 |
+
"learning_rate": 6.047661076787291e-07,
|
| 667 |
+
"loss": 1.6269,
|
| 668 |
+
"step": 106502
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 94.91642778390298,
|
| 672 |
+
"grad_norm": 1.2232398986816406,
|
| 673 |
+
"learning_rate": 5.047661076787291e-07,
|
| 674 |
+
"loss": 1.6254,
|
| 675 |
+
"step": 107635
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 95.91554575523705,
|
| 679 |
+
"grad_norm": 0.9554293751716614,
|
| 680 |
+
"learning_rate": 4.048543689320389e-07,
|
| 681 |
+
"loss": 1.6307,
|
| 682 |
+
"step": 108768
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 96.91466372657112,
|
| 686 |
+
"grad_norm": 0.42345738410949707,
|
| 687 |
+
"learning_rate": 3.0485436893203884e-07,
|
| 688 |
+
"loss": 1.6314,
|
| 689 |
+
"step": 109901
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 97.91378169790518,
|
| 693 |
+
"grad_norm": 0.4951033890247345,
|
| 694 |
+
"learning_rate": 2.0494263018534864e-07,
|
| 695 |
+
"loss": 1.6341,
|
| 696 |
+
"step": 111034
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 98.91289966923925,
|
| 700 |
+
"grad_norm": 0.40377479791641235,
|
| 701 |
+
"learning_rate": 1.0503089143865844e-07,
|
| 702 |
+
"loss": 1.6174,
|
| 703 |
+
"step": 112167
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 99.91201764057332,
|
| 707 |
+
"grad_norm": 0.5117936730384827,
|
| 708 |
+
"learning_rate": 5.03089143865843e-09,
|
| 709 |
+
"loss": 1.6313,
|
| 710 |
+
"step": 113300
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 99.91201764057332,
|
| 714 |
+
"step": 113300,
|
| 715 |
+
"total_flos": 5.364383961180322e+18,
|
| 716 |
+
"train_loss": 1.6900030501417007,
|
| 717 |
+
"train_runtime": 96835.149,
|
| 718 |
+
"train_samples_per_second": 140.49,
|
| 719 |
+
"train_steps_per_second": 1.17
|
| 720 |
+
}
|
| 721 |
+
],
|
| 722 |
+
"logging_steps": 1133,
|
| 723 |
+
"max_steps": 113300,
|
| 724 |
+
"num_input_tokens_seen": 0,
|
| 725 |
+
"num_train_epochs": 100,
|
| 726 |
+
"save_steps": 6800,
|
| 727 |
+
"stateful_callbacks": {
|
| 728 |
+
"TrainerControl": {
|
| 729 |
+
"args": {
|
| 730 |
+
"should_epoch_stop": false,
|
| 731 |
+
"should_evaluate": false,
|
| 732 |
+
"should_log": false,
|
| 733 |
+
"should_save": true,
|
| 734 |
+
"should_training_stop": true
|
| 735 |
+
},
|
| 736 |
+
"attributes": {}
|
| 737 |
+
}
|
| 738 |
+
},
|
| 739 |
+
"total_flos": 5.364383961180322e+18,
|
| 740 |
+
"train_batch_size": 30,
|
| 741 |
+
"trial_name": null,
|
| 742 |
+
"trial_params": null
|
| 743 |
+
}
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|