Commit
·
9d81a58
1
Parent(s):
6d729b4
Upload 11 files
Browse files- config.json +47 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +14 -0
- trainer_state.json +754 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
config.json
ADDED
|
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "distilbert-base-uncased",
|
| 3 |
+
"activation": "gelu",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"DistilBertForSequenceClassification"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.1,
|
| 8 |
+
"dim": 768,
|
| 9 |
+
"dropout": 0.1,
|
| 10 |
+
"hidden_dim": 3072,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "LABEL_0",
|
| 13 |
+
"1": "LABEL_1",
|
| 14 |
+
"2": "LABEL_2",
|
| 15 |
+
"3": "LABEL_3",
|
| 16 |
+
"4": "LABEL_4",
|
| 17 |
+
"5": "LABEL_5",
|
| 18 |
+
"6": "LABEL_6",
|
| 19 |
+
"7": "LABEL_7",
|
| 20 |
+
"8": "LABEL_8"
|
| 21 |
+
},
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"label2id": {
|
| 24 |
+
"LABEL_0": 0,
|
| 25 |
+
"LABEL_1": 1,
|
| 26 |
+
"LABEL_2": 2,
|
| 27 |
+
"LABEL_3": 3,
|
| 28 |
+
"LABEL_4": 4,
|
| 29 |
+
"LABEL_5": 5,
|
| 30 |
+
"LABEL_6": 6,
|
| 31 |
+
"LABEL_7": 7,
|
| 32 |
+
"LABEL_8": 8
|
| 33 |
+
},
|
| 34 |
+
"max_position_embeddings": 512,
|
| 35 |
+
"model_type": "distilbert",
|
| 36 |
+
"n_heads": 12,
|
| 37 |
+
"n_layers": 6,
|
| 38 |
+
"pad_token_id": 0,
|
| 39 |
+
"problem_type": "single_label_classification",
|
| 40 |
+
"qa_dropout": 0.1,
|
| 41 |
+
"seq_classif_dropout": 0.2,
|
| 42 |
+
"sinusoidal_pos_embds": false,
|
| 43 |
+
"tie_weights_": true,
|
| 44 |
+
"torch_dtype": "float32",
|
| 45 |
+
"transformers_version": "4.25.1",
|
| 46 |
+
"vocab_size": 30522
|
| 47 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4739baab9e5464aff0e1faaac93c35715fee05f84e9331cba182eee04309d55e
|
| 3 |
+
size 535744069
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6fa97268c4a788b2c39fba4beeec6cb4677bb81f8093d6c2a3ebd2078ce4843e
|
| 3 |
+
size 267877037
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:02416474fedc112428249a489ba86a6e9c7deaac7997f4432134532b09e85b2a
|
| 3 |
+
size 14575
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4fa665cc2fc98b6c0317c008e2683d4bf746028a7da0dd46b1f2366192ad243e
|
| 3 |
+
size 627
|
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 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"do_lower_case": true,
|
| 4 |
+
"mask_token": "[MASK]",
|
| 5 |
+
"model_max_length": 512,
|
| 6 |
+
"name_or_path": "distilbert-base-uncased",
|
| 7 |
+
"pad_token": "[PAD]",
|
| 8 |
+
"sep_token": "[SEP]",
|
| 9 |
+
"special_tokens_map_file": null,
|
| 10 |
+
"strip_accents": null,
|
| 11 |
+
"tokenize_chinese_chars": true,
|
| 12 |
+
"tokenizer_class": "DistilBertTokenizer",
|
| 13 |
+
"unk_token": "[UNK]"
|
| 14 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,754 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.021864905953407288,
|
| 3 |
+
"best_model_checkpoint": "TASK2/model1\\checkpoint-60236",
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"global_step": 60236,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.02,
|
| 12 |
+
"learning_rate": 1.983398632047281e-05,
|
| 13 |
+
"loss": 0.3368,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.03,
|
| 18 |
+
"learning_rate": 1.9667972640945615e-05,
|
| 19 |
+
"loss": 0.0613,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.05,
|
| 24 |
+
"learning_rate": 1.9501958961418424e-05,
|
| 25 |
+
"loss": 0.0552,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.07,
|
| 30 |
+
"learning_rate": 1.933594528189123e-05,
|
| 31 |
+
"loss": 0.038,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.08,
|
| 36 |
+
"learning_rate": 1.9169931602364037e-05,
|
| 37 |
+
"loss": 0.0505,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.1,
|
| 42 |
+
"learning_rate": 1.9003917922836842e-05,
|
| 43 |
+
"loss": 0.0558,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.12,
|
| 48 |
+
"learning_rate": 1.883790424330965e-05,
|
| 49 |
+
"loss": 0.0377,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.13,
|
| 54 |
+
"learning_rate": 1.867189056378246e-05,
|
| 55 |
+
"loss": 0.0452,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.15,
|
| 60 |
+
"learning_rate": 1.8505876884255264e-05,
|
| 61 |
+
"loss": 0.0395,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.17,
|
| 66 |
+
"learning_rate": 1.833986320472807e-05,
|
| 67 |
+
"loss": 0.0446,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.18,
|
| 72 |
+
"learning_rate": 1.8173849525200878e-05,
|
| 73 |
+
"loss": 0.048,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.2,
|
| 78 |
+
"learning_rate": 1.8007835845673686e-05,
|
| 79 |
+
"loss": 0.0345,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.22,
|
| 84 |
+
"learning_rate": 1.784182216614649e-05,
|
| 85 |
+
"loss": 0.0436,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.23,
|
| 90 |
+
"learning_rate": 1.76758084866193e-05,
|
| 91 |
+
"loss": 0.0486,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.25,
|
| 96 |
+
"learning_rate": 1.7509794807092108e-05,
|
| 97 |
+
"loss": 0.0529,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.27,
|
| 102 |
+
"learning_rate": 1.7343781127564913e-05,
|
| 103 |
+
"loss": 0.03,
|
| 104 |
+
"step": 8000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.28,
|
| 108 |
+
"learning_rate": 1.7177767448037718e-05,
|
| 109 |
+
"loss": 0.0445,
|
| 110 |
+
"step": 8500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.3,
|
| 114 |
+
"learning_rate": 1.7011753768510527e-05,
|
| 115 |
+
"loss": 0.0265,
|
| 116 |
+
"step": 9000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.32,
|
| 120 |
+
"learning_rate": 1.6845740088983335e-05,
|
| 121 |
+
"loss": 0.0506,
|
| 122 |
+
"step": 9500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.33,
|
| 126 |
+
"learning_rate": 1.667972640945614e-05,
|
| 127 |
+
"loss": 0.0476,
|
| 128 |
+
"step": 10000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.35,
|
| 132 |
+
"learning_rate": 1.651371272992895e-05,
|
| 133 |
+
"loss": 0.0385,
|
| 134 |
+
"step": 10500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.37,
|
| 138 |
+
"learning_rate": 1.6347699050401754e-05,
|
| 139 |
+
"loss": 0.0424,
|
| 140 |
+
"step": 11000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.38,
|
| 144 |
+
"learning_rate": 1.6181685370874562e-05,
|
| 145 |
+
"loss": 0.0259,
|
| 146 |
+
"step": 11500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.4,
|
| 150 |
+
"learning_rate": 1.6015671691347367e-05,
|
| 151 |
+
"loss": 0.0382,
|
| 152 |
+
"step": 12000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.42,
|
| 156 |
+
"learning_rate": 1.5849658011820176e-05,
|
| 157 |
+
"loss": 0.0246,
|
| 158 |
+
"step": 12500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.43,
|
| 162 |
+
"learning_rate": 1.568364433229298e-05,
|
| 163 |
+
"loss": 0.0211,
|
| 164 |
+
"step": 13000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.45,
|
| 168 |
+
"learning_rate": 1.551763065276579e-05,
|
| 169 |
+
"loss": 0.0374,
|
| 170 |
+
"step": 13500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.46,
|
| 174 |
+
"learning_rate": 1.5351616973238597e-05,
|
| 175 |
+
"loss": 0.0263,
|
| 176 |
+
"step": 14000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.48,
|
| 180 |
+
"learning_rate": 1.5185603293711404e-05,
|
| 181 |
+
"loss": 0.0396,
|
| 182 |
+
"step": 14500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.5,
|
| 186 |
+
"learning_rate": 1.501958961418421e-05,
|
| 187 |
+
"loss": 0.0275,
|
| 188 |
+
"step": 15000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.51,
|
| 192 |
+
"learning_rate": 1.4853575934657016e-05,
|
| 193 |
+
"loss": 0.0251,
|
| 194 |
+
"step": 15500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.53,
|
| 198 |
+
"learning_rate": 1.4687562255129823e-05,
|
| 199 |
+
"loss": 0.0368,
|
| 200 |
+
"step": 16000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.55,
|
| 204 |
+
"learning_rate": 1.4521548575602631e-05,
|
| 205 |
+
"loss": 0.0322,
|
| 206 |
+
"step": 16500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.56,
|
| 210 |
+
"learning_rate": 1.4355534896075438e-05,
|
| 211 |
+
"loss": 0.0285,
|
| 212 |
+
"step": 17000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.58,
|
| 216 |
+
"learning_rate": 1.4189521216548245e-05,
|
| 217 |
+
"loss": 0.0298,
|
| 218 |
+
"step": 17500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.6,
|
| 222 |
+
"learning_rate": 1.4023507537021053e-05,
|
| 223 |
+
"loss": 0.0425,
|
| 224 |
+
"step": 18000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.61,
|
| 228 |
+
"learning_rate": 1.3857493857493858e-05,
|
| 229 |
+
"loss": 0.0307,
|
| 230 |
+
"step": 18500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.63,
|
| 234 |
+
"learning_rate": 1.3691480177966665e-05,
|
| 235 |
+
"loss": 0.0233,
|
| 236 |
+
"step": 19000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.65,
|
| 240 |
+
"learning_rate": 1.3525466498439472e-05,
|
| 241 |
+
"loss": 0.0246,
|
| 242 |
+
"step": 19500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.66,
|
| 246 |
+
"learning_rate": 1.335945281891228e-05,
|
| 247 |
+
"loss": 0.0315,
|
| 248 |
+
"step": 20000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.68,
|
| 252 |
+
"learning_rate": 1.3193439139385087e-05,
|
| 253 |
+
"loss": 0.0315,
|
| 254 |
+
"step": 20500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.7,
|
| 258 |
+
"learning_rate": 1.3027425459857894e-05,
|
| 259 |
+
"loss": 0.0329,
|
| 260 |
+
"step": 21000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.71,
|
| 264 |
+
"learning_rate": 1.2861411780330699e-05,
|
| 265 |
+
"loss": 0.0258,
|
| 266 |
+
"step": 21500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.73,
|
| 270 |
+
"learning_rate": 1.2695398100803507e-05,
|
| 271 |
+
"loss": 0.0274,
|
| 272 |
+
"step": 22000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.75,
|
| 276 |
+
"learning_rate": 1.2529384421276314e-05,
|
| 277 |
+
"loss": 0.0273,
|
| 278 |
+
"step": 22500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.76,
|
| 282 |
+
"learning_rate": 1.236337074174912e-05,
|
| 283 |
+
"loss": 0.0309,
|
| 284 |
+
"step": 23000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.78,
|
| 288 |
+
"learning_rate": 1.219735706222193e-05,
|
| 289 |
+
"loss": 0.0255,
|
| 290 |
+
"step": 23500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.8,
|
| 294 |
+
"learning_rate": 1.2031343382694736e-05,
|
| 295 |
+
"loss": 0.0331,
|
| 296 |
+
"step": 24000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.81,
|
| 300 |
+
"learning_rate": 1.1865329703167543e-05,
|
| 301 |
+
"loss": 0.028,
|
| 302 |
+
"step": 24500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.83,
|
| 306 |
+
"learning_rate": 1.1699316023640348e-05,
|
| 307 |
+
"loss": 0.0273,
|
| 308 |
+
"step": 25000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.85,
|
| 312 |
+
"learning_rate": 1.1533302344113154e-05,
|
| 313 |
+
"loss": 0.0222,
|
| 314 |
+
"step": 25500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.86,
|
| 318 |
+
"learning_rate": 1.1367288664585963e-05,
|
| 319 |
+
"loss": 0.0308,
|
| 320 |
+
"step": 26000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.88,
|
| 324 |
+
"learning_rate": 1.120127498505877e-05,
|
| 325 |
+
"loss": 0.0363,
|
| 326 |
+
"step": 26500
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.9,
|
| 330 |
+
"learning_rate": 1.1035261305531576e-05,
|
| 331 |
+
"loss": 0.0324,
|
| 332 |
+
"step": 27000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.91,
|
| 336 |
+
"learning_rate": 1.0869247626004385e-05,
|
| 337 |
+
"loss": 0.0221,
|
| 338 |
+
"step": 27500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.93,
|
| 342 |
+
"learning_rate": 1.070323394647719e-05,
|
| 343 |
+
"loss": 0.0199,
|
| 344 |
+
"step": 28000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.95,
|
| 348 |
+
"learning_rate": 1.0537220266949997e-05,
|
| 349 |
+
"loss": 0.0256,
|
| 350 |
+
"step": 28500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.96,
|
| 354 |
+
"learning_rate": 1.0371206587422803e-05,
|
| 355 |
+
"loss": 0.0294,
|
| 356 |
+
"step": 29000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.98,
|
| 360 |
+
"learning_rate": 1.0205192907895612e-05,
|
| 361 |
+
"loss": 0.0254,
|
| 362 |
+
"step": 29500
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 1.0,
|
| 366 |
+
"learning_rate": 1.0039179228368419e-05,
|
| 367 |
+
"loss": 0.0314,
|
| 368 |
+
"step": 30000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 1.0,
|
| 372 |
+
"eval_accuracy": 0.9955094252722309,
|
| 373 |
+
"eval_loss": 0.023077525198459625,
|
| 374 |
+
"eval_runtime": 2150.2467,
|
| 375 |
+
"eval_samples_per_second": 28.273,
|
| 376 |
+
"eval_steps_per_second": 3.534,
|
| 377 |
+
"step": 30118
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 1.01,
|
| 381 |
+
"learning_rate": 9.873165548841225e-06,
|
| 382 |
+
"loss": 0.0142,
|
| 383 |
+
"step": 30500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 1.03,
|
| 387 |
+
"learning_rate": 9.707151869314032e-06,
|
| 388 |
+
"loss": 0.0214,
|
| 389 |
+
"step": 31000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.05,
|
| 393 |
+
"learning_rate": 9.541138189786839e-06,
|
| 394 |
+
"loss": 0.0092,
|
| 395 |
+
"step": 31500
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.06,
|
| 399 |
+
"learning_rate": 9.375124510259647e-06,
|
| 400 |
+
"loss": 0.0124,
|
| 401 |
+
"step": 32000
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.08,
|
| 405 |
+
"learning_rate": 9.209110830732452e-06,
|
| 406 |
+
"loss": 0.0151,
|
| 407 |
+
"step": 32500
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 1.1,
|
| 411 |
+
"learning_rate": 9.043097151205261e-06,
|
| 412 |
+
"loss": 0.012,
|
| 413 |
+
"step": 33000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 1.11,
|
| 417 |
+
"learning_rate": 8.877083471678068e-06,
|
| 418 |
+
"loss": 0.017,
|
| 419 |
+
"step": 33500
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 1.13,
|
| 423 |
+
"learning_rate": 8.711069792150874e-06,
|
| 424 |
+
"loss": 0.0205,
|
| 425 |
+
"step": 34000
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 1.15,
|
| 429 |
+
"learning_rate": 8.545056112623681e-06,
|
| 430 |
+
"loss": 0.0205,
|
| 431 |
+
"step": 34500
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 1.16,
|
| 435 |
+
"learning_rate": 8.379042433096488e-06,
|
| 436 |
+
"loss": 0.0205,
|
| 437 |
+
"step": 35000
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.18,
|
| 441 |
+
"learning_rate": 8.213028753569295e-06,
|
| 442 |
+
"loss": 0.0123,
|
| 443 |
+
"step": 35500
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.2,
|
| 447 |
+
"learning_rate": 8.047015074042101e-06,
|
| 448 |
+
"loss": 0.0149,
|
| 449 |
+
"step": 36000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 1.21,
|
| 453 |
+
"learning_rate": 7.881001394514908e-06,
|
| 454 |
+
"loss": 0.0142,
|
| 455 |
+
"step": 36500
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 1.23,
|
| 459 |
+
"learning_rate": 7.714987714987717e-06,
|
| 460 |
+
"loss": 0.0148,
|
| 461 |
+
"step": 37000
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 1.25,
|
| 465 |
+
"learning_rate": 7.5489740354605225e-06,
|
| 466 |
+
"loss": 0.0093,
|
| 467 |
+
"step": 37500
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 1.26,
|
| 471 |
+
"learning_rate": 7.382960355933329e-06,
|
| 472 |
+
"loss": 0.0111,
|
| 473 |
+
"step": 38000
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 1.28,
|
| 477 |
+
"learning_rate": 7.216946676406137e-06,
|
| 478 |
+
"loss": 0.0163,
|
| 479 |
+
"step": 38500
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 1.29,
|
| 483 |
+
"learning_rate": 7.050932996878943e-06,
|
| 484 |
+
"loss": 0.0198,
|
| 485 |
+
"step": 39000
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.31,
|
| 489 |
+
"learning_rate": 6.88491931735175e-06,
|
| 490 |
+
"loss": 0.0146,
|
| 491 |
+
"step": 39500
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 1.33,
|
| 495 |
+
"learning_rate": 6.718905637824558e-06,
|
| 496 |
+
"loss": 0.0252,
|
| 497 |
+
"step": 40000
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 1.34,
|
| 501 |
+
"learning_rate": 6.552891958297364e-06,
|
| 502 |
+
"loss": 0.0242,
|
| 503 |
+
"step": 40500
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 1.36,
|
| 507 |
+
"learning_rate": 6.3868782787701715e-06,
|
| 508 |
+
"loss": 0.0096,
|
| 509 |
+
"step": 41000
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 1.38,
|
| 513 |
+
"learning_rate": 6.220864599242978e-06,
|
| 514 |
+
"loss": 0.0193,
|
| 515 |
+
"step": 41500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 1.39,
|
| 519 |
+
"learning_rate": 6.054850919715784e-06,
|
| 520 |
+
"loss": 0.0143,
|
| 521 |
+
"step": 42000
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 1.41,
|
| 525 |
+
"learning_rate": 5.888837240188592e-06,
|
| 526 |
+
"loss": 0.0156,
|
| 527 |
+
"step": 42500
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.43,
|
| 531 |
+
"learning_rate": 5.722823560661399e-06,
|
| 532 |
+
"loss": 0.0129,
|
| 533 |
+
"step": 43000
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 1.44,
|
| 537 |
+
"learning_rate": 5.556809881134206e-06,
|
| 538 |
+
"loss": 0.0151,
|
| 539 |
+
"step": 43500
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 1.46,
|
| 543 |
+
"learning_rate": 5.390796201607013e-06,
|
| 544 |
+
"loss": 0.0063,
|
| 545 |
+
"step": 44000
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 1.48,
|
| 549 |
+
"learning_rate": 5.2247825220798196e-06,
|
| 550 |
+
"loss": 0.0137,
|
| 551 |
+
"step": 44500
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 1.49,
|
| 555 |
+
"learning_rate": 5.058768842552627e-06,
|
| 556 |
+
"loss": 0.0108,
|
| 557 |
+
"step": 45000
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 1.51,
|
| 561 |
+
"learning_rate": 4.892755163025434e-06,
|
| 562 |
+
"loss": 0.0212,
|
| 563 |
+
"step": 45500
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 1.53,
|
| 567 |
+
"learning_rate": 4.726741483498241e-06,
|
| 568 |
+
"loss": 0.0111,
|
| 569 |
+
"step": 46000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.54,
|
| 573 |
+
"learning_rate": 4.560727803971048e-06,
|
| 574 |
+
"loss": 0.011,
|
| 575 |
+
"step": 46500
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 1.56,
|
| 579 |
+
"learning_rate": 4.394714124443854e-06,
|
| 580 |
+
"loss": 0.007,
|
| 581 |
+
"step": 47000
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 1.58,
|
| 585 |
+
"learning_rate": 4.228700444916661e-06,
|
| 586 |
+
"loss": 0.0117,
|
| 587 |
+
"step": 47500
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 1.59,
|
| 591 |
+
"learning_rate": 4.0626867653894685e-06,
|
| 592 |
+
"loss": 0.0078,
|
| 593 |
+
"step": 48000
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 1.61,
|
| 597 |
+
"learning_rate": 3.896673085862275e-06,
|
| 598 |
+
"loss": 0.0058,
|
| 599 |
+
"step": 48500
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 1.63,
|
| 603 |
+
"learning_rate": 3.730659406335082e-06,
|
| 604 |
+
"loss": 0.0133,
|
| 605 |
+
"step": 49000
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 1.64,
|
| 609 |
+
"learning_rate": 3.5646457268078892e-06,
|
| 610 |
+
"loss": 0.0097,
|
| 611 |
+
"step": 49500
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.66,
|
| 615 |
+
"learning_rate": 3.398632047280696e-06,
|
| 616 |
+
"loss": 0.0111,
|
| 617 |
+
"step": 50000
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 1.68,
|
| 621 |
+
"learning_rate": 3.232618367753503e-06,
|
| 622 |
+
"loss": 0.016,
|
| 623 |
+
"step": 50500
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 1.69,
|
| 627 |
+
"learning_rate": 3.06660468822631e-06,
|
| 628 |
+
"loss": 0.0147,
|
| 629 |
+
"step": 51000
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 1.71,
|
| 633 |
+
"learning_rate": 2.9005910086991167e-06,
|
| 634 |
+
"loss": 0.0112,
|
| 635 |
+
"step": 51500
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 1.73,
|
| 639 |
+
"learning_rate": 2.7345773291719243e-06,
|
| 640 |
+
"loss": 0.0077,
|
| 641 |
+
"step": 52000
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 1.74,
|
| 645 |
+
"learning_rate": 2.568563649644731e-06,
|
| 646 |
+
"loss": 0.0123,
|
| 647 |
+
"step": 52500
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.76,
|
| 651 |
+
"learning_rate": 2.4025499701175378e-06,
|
| 652 |
+
"loss": 0.0102,
|
| 653 |
+
"step": 53000
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.78,
|
| 657 |
+
"learning_rate": 2.236536290590345e-06,
|
| 658 |
+
"loss": 0.0109,
|
| 659 |
+
"step": 53500
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 1.79,
|
| 663 |
+
"learning_rate": 2.0705226110631517e-06,
|
| 664 |
+
"loss": 0.01,
|
| 665 |
+
"step": 54000
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 1.81,
|
| 669 |
+
"learning_rate": 1.9045089315359589e-06,
|
| 670 |
+
"loss": 0.0161,
|
| 671 |
+
"step": 54500
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 1.83,
|
| 675 |
+
"learning_rate": 1.7384952520087656e-06,
|
| 676 |
+
"loss": 0.0104,
|
| 677 |
+
"step": 55000
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 1.84,
|
| 681 |
+
"learning_rate": 1.5724815724815726e-06,
|
| 682 |
+
"loss": 0.008,
|
| 683 |
+
"step": 55500
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 1.86,
|
| 687 |
+
"learning_rate": 1.4064678929543795e-06,
|
| 688 |
+
"loss": 0.0101,
|
| 689 |
+
"step": 56000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.88,
|
| 693 |
+
"learning_rate": 1.2404542134271865e-06,
|
| 694 |
+
"loss": 0.0093,
|
| 695 |
+
"step": 56500
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.89,
|
| 699 |
+
"learning_rate": 1.0744405338999935e-06,
|
| 700 |
+
"loss": 0.0155,
|
| 701 |
+
"step": 57000
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 1.91,
|
| 705 |
+
"learning_rate": 9.084268543728004e-07,
|
| 706 |
+
"loss": 0.0109,
|
| 707 |
+
"step": 57500
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 1.93,
|
| 711 |
+
"learning_rate": 7.424131748456073e-07,
|
| 712 |
+
"loss": 0.008,
|
| 713 |
+
"step": 58000
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 1.94,
|
| 717 |
+
"learning_rate": 5.763994953184143e-07,
|
| 718 |
+
"loss": 0.0108,
|
| 719 |
+
"step": 58500
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 1.96,
|
| 723 |
+
"learning_rate": 4.103858157912213e-07,
|
| 724 |
+
"loss": 0.0155,
|
| 725 |
+
"step": 59000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 1.98,
|
| 729 |
+
"learning_rate": 2.4437213626402814e-07,
|
| 730 |
+
"loss": 0.0079,
|
| 731 |
+
"step": 59500
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 1.99,
|
| 735 |
+
"learning_rate": 7.835845673683512e-08,
|
| 736 |
+
"loss": 0.0113,
|
| 737 |
+
"step": 60000
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 2.0,
|
| 741 |
+
"eval_accuracy": 0.9959042010724742,
|
| 742 |
+
"eval_loss": 0.021864905953407288,
|
| 743 |
+
"eval_runtime": 2302.6637,
|
| 744 |
+
"eval_samples_per_second": 26.402,
|
| 745 |
+
"eval_steps_per_second": 3.301,
|
| 746 |
+
"step": 60236
|
| 747 |
+
}
|
| 748 |
+
],
|
| 749 |
+
"max_steps": 60236,
|
| 750 |
+
"num_train_epochs": 2,
|
| 751 |
+
"total_flos": 6.384188864533709e+16,
|
| 752 |
+
"trial_name": null,
|
| 753 |
+
"trial_params": null
|
| 754 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:02c492007729786fe5443193c8bd7169e30b90b368b878cd341d00486775d12e
|
| 3 |
+
size 3387
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|