model : 0523013301
Browse files- config.json +69 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +55 -0
- trainer_state.json +1686 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
config.json
ADDED
|
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "distilbert/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": "Jersey Basic",
|
| 13 |
+
"1": "Under-, Nightwear",
|
| 14 |
+
"2": "Socks and Tights",
|
| 15 |
+
"3": "Jersey Fancy",
|
| 16 |
+
"4": "Accessories",
|
| 17 |
+
"5": "Trousers Denim",
|
| 18 |
+
"6": "Outdoor",
|
| 19 |
+
"7": "Shoes",
|
| 20 |
+
"8": "Swimwear",
|
| 21 |
+
"9": "Knitwear",
|
| 22 |
+
"10": "Shirts",
|
| 23 |
+
"11": "Trousers",
|
| 24 |
+
"12": "Dressed",
|
| 25 |
+
"13": "Shorts",
|
| 26 |
+
"14": "Dresses Ladies",
|
| 27 |
+
"15": "Skirts",
|
| 28 |
+
"16": "Special Offers",
|
| 29 |
+
"17": "Blouses",
|
| 30 |
+
"18": "Woven/Jersey/Knitted mix Baby",
|
| 31 |
+
"19": "Dresses/Skirts girls"
|
| 32 |
+
},
|
| 33 |
+
"initializer_range": 0.02,
|
| 34 |
+
"label2id": {
|
| 35 |
+
"Accessories": 4,
|
| 36 |
+
"Blouses": 17,
|
| 37 |
+
"Dressed": 12,
|
| 38 |
+
"Dresses Ladies": 14,
|
| 39 |
+
"Dresses/Skirts girls": 19,
|
| 40 |
+
"Jersey Basic": 0,
|
| 41 |
+
"Jersey Fancy": 3,
|
| 42 |
+
"Knitwear": 9,
|
| 43 |
+
"Outdoor": 6,
|
| 44 |
+
"Shirts": 10,
|
| 45 |
+
"Shoes": 7,
|
| 46 |
+
"Shorts": 13,
|
| 47 |
+
"Skirts": 15,
|
| 48 |
+
"Socks and Tights": 2,
|
| 49 |
+
"Special Offers": 16,
|
| 50 |
+
"Swimwear": 8,
|
| 51 |
+
"Trousers": 11,
|
| 52 |
+
"Trousers Denim": 5,
|
| 53 |
+
"Under-, Nightwear": 1,
|
| 54 |
+
"Woven/Jersey/Knitted mix Baby": 18
|
| 55 |
+
},
|
| 56 |
+
"max_position_embeddings": 512,
|
| 57 |
+
"model_type": "distilbert",
|
| 58 |
+
"n_heads": 12,
|
| 59 |
+
"n_layers": 6,
|
| 60 |
+
"pad_token_id": 0,
|
| 61 |
+
"problem_type": "single_label_classification",
|
| 62 |
+
"qa_dropout": 0.1,
|
| 63 |
+
"seq_classif_dropout": 0.2,
|
| 64 |
+
"sinusoidal_pos_embds": false,
|
| 65 |
+
"tie_weights_": true,
|
| 66 |
+
"torch_dtype": "float32",
|
| 67 |
+
"transformers_version": "4.41.0",
|
| 68 |
+
"vocab_size": 30522
|
| 69 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:291f893940f0965009475392375fa0456885a46f72341e28a479d4c3ae1e90fa
|
| 3 |
+
size 267887936
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:675207e42cb9771234a18cd05e4157e1010d362a8c9da9b6abb3d19bb9a54f8e
|
| 3 |
+
size 535838010
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:76d2b21de1f75cf4cfeaae79e1d9149037272d522e6a1fd7c551b4028a9f7e25
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7af99a8771cbf6f19faee7183ef06b0cb12033a679e66ad2f10dc3b2030aef2d
|
| 3 |
+
size 1064
|
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 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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_lower_case": true,
|
| 47 |
+
"mask_token": "[MASK]",
|
| 48 |
+
"model_max_length": 512,
|
| 49 |
+
"pad_token": "[PAD]",
|
| 50 |
+
"sep_token": "[SEP]",
|
| 51 |
+
"strip_accents": null,
|
| 52 |
+
"tokenize_chinese_chars": true,
|
| 53 |
+
"tokenizer_class": "DistilBertTokenizer",
|
| 54 |
+
"unk_token": "[UNK]"
|
| 55 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1686 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.31098222732543945,
|
| 3 |
+
"best_model_checkpoint": "./seq_clf/0523013301/checkpoint-24528",
|
| 4 |
+
"epoch": 25.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 102200,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.1223091976516634,
|
| 13 |
+
"grad_norm": 3.0948333740234375,
|
| 14 |
+
"learning_rate": 1.990215264187867e-05,
|
| 15 |
+
"loss": 1.4987,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.2446183953033268,
|
| 20 |
+
"grad_norm": 4.519032955169678,
|
| 21 |
+
"learning_rate": 1.9804305283757338e-05,
|
| 22 |
+
"loss": 0.8512,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.3669275929549902,
|
| 27 |
+
"grad_norm": 6.118898868560791,
|
| 28 |
+
"learning_rate": 1.970645792563601e-05,
|
| 29 |
+
"loss": 0.7319,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.4892367906066536,
|
| 34 |
+
"grad_norm": 5.870689392089844,
|
| 35 |
+
"learning_rate": 1.9608610567514678e-05,
|
| 36 |
+
"loss": 0.673,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.6115459882583171,
|
| 41 |
+
"grad_norm": 5.046342849731445,
|
| 42 |
+
"learning_rate": 1.951076320939335e-05,
|
| 43 |
+
"loss": 0.6408,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.7338551859099804,
|
| 48 |
+
"grad_norm": 5.137794494628906,
|
| 49 |
+
"learning_rate": 1.9412915851272015e-05,
|
| 50 |
+
"loss": 0.6029,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.8561643835616438,
|
| 55 |
+
"grad_norm": 5.616779327392578,
|
| 56 |
+
"learning_rate": 1.9315068493150686e-05,
|
| 57 |
+
"loss": 0.5867,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.9784735812133072,
|
| 62 |
+
"grad_norm": 3.4356954097747803,
|
| 63 |
+
"learning_rate": 1.9217221135029354e-05,
|
| 64 |
+
"loss": 0.5654,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 1.0,
|
| 69 |
+
"eval_accuracy": 0.8249802208386364,
|
| 70 |
+
"eval_loss": 0.5318922400474548,
|
| 71 |
+
"eval_runtime": 77.9312,
|
| 72 |
+
"eval_samples_per_second": 373.034,
|
| 73 |
+
"eval_steps_per_second": 11.664,
|
| 74 |
+
"step": 4088
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.1007827788649707,
|
| 78 |
+
"grad_norm": 5.042200088500977,
|
| 79 |
+
"learning_rate": 1.9119373776908026e-05,
|
| 80 |
+
"loss": 0.5171,
|
| 81 |
+
"step": 4500
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.223091976516634,
|
| 85 |
+
"grad_norm": 5.630020618438721,
|
| 86 |
+
"learning_rate": 1.9021526418786694e-05,
|
| 87 |
+
"loss": 0.4983,
|
| 88 |
+
"step": 5000
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.3454011741682974,
|
| 92 |
+
"grad_norm": 3.5620667934417725,
|
| 93 |
+
"learning_rate": 1.8923679060665363e-05,
|
| 94 |
+
"loss": 0.488,
|
| 95 |
+
"step": 5500
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.467710371819961,
|
| 99 |
+
"grad_norm": 6.017724990844727,
|
| 100 |
+
"learning_rate": 1.8825831702544034e-05,
|
| 101 |
+
"loss": 0.4867,
|
| 102 |
+
"step": 6000
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.5900195694716244,
|
| 106 |
+
"grad_norm": 4.974002361297607,
|
| 107 |
+
"learning_rate": 1.8727984344422703e-05,
|
| 108 |
+
"loss": 0.4645,
|
| 109 |
+
"step": 6500
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.7123287671232876,
|
| 113 |
+
"grad_norm": 5.986278057098389,
|
| 114 |
+
"learning_rate": 1.863013698630137e-05,
|
| 115 |
+
"loss": 0.4582,
|
| 116 |
+
"step": 7000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.8346379647749511,
|
| 120 |
+
"grad_norm": 5.255044937133789,
|
| 121 |
+
"learning_rate": 1.853228962818004e-05,
|
| 122 |
+
"loss": 0.4609,
|
| 123 |
+
"step": 7500
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.9569471624266144,
|
| 127 |
+
"grad_norm": 5.023875713348389,
|
| 128 |
+
"learning_rate": 1.843444227005871e-05,
|
| 129 |
+
"loss": 0.4495,
|
| 130 |
+
"step": 8000
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 2.0,
|
| 134 |
+
"eval_accuracy": 0.8506759313405111,
|
| 135 |
+
"eval_loss": 0.4426937699317932,
|
| 136 |
+
"eval_runtime": 78.0987,
|
| 137 |
+
"eval_samples_per_second": 372.234,
|
| 138 |
+
"eval_steps_per_second": 11.639,
|
| 139 |
+
"step": 8176
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 2.079256360078278,
|
| 143 |
+
"grad_norm": 6.947430610656738,
|
| 144 |
+
"learning_rate": 1.833659491193738e-05,
|
| 145 |
+
"loss": 0.4058,
|
| 146 |
+
"step": 8500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 2.2015655577299413,
|
| 150 |
+
"grad_norm": 2.9549953937530518,
|
| 151 |
+
"learning_rate": 1.8238747553816047e-05,
|
| 152 |
+
"loss": 0.3764,
|
| 153 |
+
"step": 9000
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.323874755381605,
|
| 157 |
+
"grad_norm": 6.063886642456055,
|
| 158 |
+
"learning_rate": 1.814090019569472e-05,
|
| 159 |
+
"loss": 0.3785,
|
| 160 |
+
"step": 9500
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 2.446183953033268,
|
| 164 |
+
"grad_norm": 6.911349296569824,
|
| 165 |
+
"learning_rate": 1.8043052837573387e-05,
|
| 166 |
+
"loss": 0.3736,
|
| 167 |
+
"step": 10000
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 2.5684931506849313,
|
| 171 |
+
"grad_norm": 5.118221282958984,
|
| 172 |
+
"learning_rate": 1.7945205479452055e-05,
|
| 173 |
+
"loss": 0.3684,
|
| 174 |
+
"step": 10500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 2.690802348336595,
|
| 178 |
+
"grad_norm": 4.369062423706055,
|
| 179 |
+
"learning_rate": 1.7847358121330724e-05,
|
| 180 |
+
"loss": 0.3607,
|
| 181 |
+
"step": 11000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 2.8131115459882583,
|
| 185 |
+
"grad_norm": 6.040197849273682,
|
| 186 |
+
"learning_rate": 1.7749510763209395e-05,
|
| 187 |
+
"loss": 0.3657,
|
| 188 |
+
"step": 11500
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.935420743639922,
|
| 192 |
+
"grad_norm": 6.751057147979736,
|
| 193 |
+
"learning_rate": 1.7651663405088064e-05,
|
| 194 |
+
"loss": 0.3517,
|
| 195 |
+
"step": 12000
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 3.0,
|
| 199 |
+
"eval_accuracy": 0.8710398679096006,
|
| 200 |
+
"eval_loss": 0.3752569854259491,
|
| 201 |
+
"eval_runtime": 78.0858,
|
| 202 |
+
"eval_samples_per_second": 372.295,
|
| 203 |
+
"eval_steps_per_second": 11.641,
|
| 204 |
+
"step": 12264
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 3.0577299412915853,
|
| 208 |
+
"grad_norm": 5.28397798538208,
|
| 209 |
+
"learning_rate": 1.7553816046966735e-05,
|
| 210 |
+
"loss": 0.3172,
|
| 211 |
+
"step": 12500
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 3.1800391389432487,
|
| 215 |
+
"grad_norm": 3.873892068862915,
|
| 216 |
+
"learning_rate": 1.74559686888454e-05,
|
| 217 |
+
"loss": 0.2865,
|
| 218 |
+
"step": 13000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 3.302348336594912,
|
| 222 |
+
"grad_norm": 3.584193468093872,
|
| 223 |
+
"learning_rate": 1.735812133072407e-05,
|
| 224 |
+
"loss": 0.285,
|
| 225 |
+
"step": 13500
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 3.4246575342465753,
|
| 229 |
+
"grad_norm": 6.080500602722168,
|
| 230 |
+
"learning_rate": 1.726027397260274e-05,
|
| 231 |
+
"loss": 0.2867,
|
| 232 |
+
"step": 14000
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 3.5469667318982387,
|
| 236 |
+
"grad_norm": 5.05246114730835,
|
| 237 |
+
"learning_rate": 1.716242661448141e-05,
|
| 238 |
+
"loss": 0.2793,
|
| 239 |
+
"step": 14500
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 3.6692759295499022,
|
| 243 |
+
"grad_norm": 4.842950344085693,
|
| 244 |
+
"learning_rate": 1.706457925636008e-05,
|
| 245 |
+
"loss": 0.2829,
|
| 246 |
+
"step": 15000
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 3.7915851272015657,
|
| 250 |
+
"grad_norm": 6.52248477935791,
|
| 251 |
+
"learning_rate": 1.6966731898238748e-05,
|
| 252 |
+
"loss": 0.2776,
|
| 253 |
+
"step": 15500
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 3.9138943248532287,
|
| 257 |
+
"grad_norm": 5.814143657684326,
|
| 258 |
+
"learning_rate": 1.686888454011742e-05,
|
| 259 |
+
"loss": 0.268,
|
| 260 |
+
"step": 16000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 4.0,
|
| 264 |
+
"eval_accuracy": 0.8836985311822779,
|
| 265 |
+
"eval_loss": 0.33874690532684326,
|
| 266 |
+
"eval_runtime": 78.7296,
|
| 267 |
+
"eval_samples_per_second": 369.251,
|
| 268 |
+
"eval_steps_per_second": 11.546,
|
| 269 |
+
"step": 16352
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 4.036203522504892,
|
| 273 |
+
"grad_norm": 4.4718194007873535,
|
| 274 |
+
"learning_rate": 1.6771037181996088e-05,
|
| 275 |
+
"loss": 0.2524,
|
| 276 |
+
"step": 16500
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 4.158512720156556,
|
| 280 |
+
"grad_norm": 6.8225531578063965,
|
| 281 |
+
"learning_rate": 1.6673189823874756e-05,
|
| 282 |
+
"loss": 0.2121,
|
| 283 |
+
"step": 17000
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 4.280821917808219,
|
| 287 |
+
"grad_norm": 4.292703151702881,
|
| 288 |
+
"learning_rate": 1.6575342465753425e-05,
|
| 289 |
+
"loss": 0.2168,
|
| 290 |
+
"step": 17500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 4.403131115459883,
|
| 294 |
+
"grad_norm": 6.740828514099121,
|
| 295 |
+
"learning_rate": 1.6477495107632096e-05,
|
| 296 |
+
"loss": 0.2079,
|
| 297 |
+
"step": 18000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 4.525440313111546,
|
| 301 |
+
"grad_norm": 7.279580593109131,
|
| 302 |
+
"learning_rate": 1.6379647749510764e-05,
|
| 303 |
+
"loss": 0.2178,
|
| 304 |
+
"step": 18500
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 4.64774951076321,
|
| 308 |
+
"grad_norm": 8.839160919189453,
|
| 309 |
+
"learning_rate": 1.6281800391389433e-05,
|
| 310 |
+
"loss": 0.2135,
|
| 311 |
+
"step": 19000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 4.770058708414873,
|
| 315 |
+
"grad_norm": 5.748088836669922,
|
| 316 |
+
"learning_rate": 1.6183953033268104e-05,
|
| 317 |
+
"loss": 0.215,
|
| 318 |
+
"step": 19500
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 4.892367906066536,
|
| 322 |
+
"grad_norm": 6.222439765930176,
|
| 323 |
+
"learning_rate": 1.6086105675146773e-05,
|
| 324 |
+
"loss": 0.2114,
|
| 325 |
+
"step": 20000
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 5.0,
|
| 329 |
+
"eval_accuracy": 0.895531629458911,
|
| 330 |
+
"eval_loss": 0.3136844038963318,
|
| 331 |
+
"eval_runtime": 78.002,
|
| 332 |
+
"eval_samples_per_second": 372.695,
|
| 333 |
+
"eval_steps_per_second": 11.654,
|
| 334 |
+
"step": 20440
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 5.014677103718199,
|
| 338 |
+
"grad_norm": 5.067689895629883,
|
| 339 |
+
"learning_rate": 1.598825831702544e-05,
|
| 340 |
+
"loss": 0.2001,
|
| 341 |
+
"step": 20500
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 5.136986301369863,
|
| 345 |
+
"grad_norm": 5.334500789642334,
|
| 346 |
+
"learning_rate": 1.589041095890411e-05,
|
| 347 |
+
"loss": 0.1574,
|
| 348 |
+
"step": 21000
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 5.259295499021526,
|
| 352 |
+
"grad_norm": 8.226490020751953,
|
| 353 |
+
"learning_rate": 1.579256360078278e-05,
|
| 354 |
+
"loss": 0.1599,
|
| 355 |
+
"step": 21500
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 5.38160469667319,
|
| 359 |
+
"grad_norm": 5.279560565948486,
|
| 360 |
+
"learning_rate": 1.569471624266145e-05,
|
| 361 |
+
"loss": 0.1594,
|
| 362 |
+
"step": 22000
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 5.503913894324853,
|
| 366 |
+
"grad_norm": 8.88842487335205,
|
| 367 |
+
"learning_rate": 1.559686888454012e-05,
|
| 368 |
+
"loss": 0.1577,
|
| 369 |
+
"step": 22500
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 5.626223091976517,
|
| 373 |
+
"grad_norm": 6.64274787902832,
|
| 374 |
+
"learning_rate": 1.5499021526418785e-05,
|
| 375 |
+
"loss": 0.1566,
|
| 376 |
+
"step": 23000
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 5.74853228962818,
|
| 380 |
+
"grad_norm": 8.864998817443848,
|
| 381 |
+
"learning_rate": 1.5401174168297457e-05,
|
| 382 |
+
"loss": 0.163,
|
| 383 |
+
"step": 23500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 5.870841487279844,
|
| 387 |
+
"grad_norm": 4.494806289672852,
|
| 388 |
+
"learning_rate": 1.5303326810176125e-05,
|
| 389 |
+
"loss": 0.159,
|
| 390 |
+
"step": 24000
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 5.993150684931507,
|
| 394 |
+
"grad_norm": 8.841043472290039,
|
| 395 |
+
"learning_rate": 1.5205479452054797e-05,
|
| 396 |
+
"loss": 0.1593,
|
| 397 |
+
"step": 24500
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 6.0,
|
| 401 |
+
"eval_accuracy": 0.9032025042138213,
|
| 402 |
+
"eval_loss": 0.31098222732543945,
|
| 403 |
+
"eval_runtime": 78.1256,
|
| 404 |
+
"eval_samples_per_second": 372.106,
|
| 405 |
+
"eval_steps_per_second": 11.635,
|
| 406 |
+
"step": 24528
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 6.1154598825831705,
|
| 410 |
+
"grad_norm": 5.516913890838623,
|
| 411 |
+
"learning_rate": 1.5107632093933464e-05,
|
| 412 |
+
"loss": 0.1186,
|
| 413 |
+
"step": 25000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 6.237769080234834,
|
| 417 |
+
"grad_norm": 17.19109344482422,
|
| 418 |
+
"learning_rate": 1.5009784735812134e-05,
|
| 419 |
+
"loss": 0.1242,
|
| 420 |
+
"step": 25500
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 6.3600782778864975,
|
| 424 |
+
"grad_norm": 4.817780494689941,
|
| 425 |
+
"learning_rate": 1.4911937377690804e-05,
|
| 426 |
+
"loss": 0.1264,
|
| 427 |
+
"step": 26000
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 6.48238747553816,
|
| 431 |
+
"grad_norm": 9.93541431427002,
|
| 432 |
+
"learning_rate": 1.4814090019569473e-05,
|
| 433 |
+
"loss": 0.1251,
|
| 434 |
+
"step": 26500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 6.604696673189824,
|
| 438 |
+
"grad_norm": 7.402759552001953,
|
| 439 |
+
"learning_rate": 1.4716242661448142e-05,
|
| 440 |
+
"loss": 0.1205,
|
| 441 |
+
"step": 27000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 6.727005870841487,
|
| 445 |
+
"grad_norm": 4.997498989105225,
|
| 446 |
+
"learning_rate": 1.4618395303326812e-05,
|
| 447 |
+
"loss": 0.1221,
|
| 448 |
+
"step": 27500
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 6.8493150684931505,
|
| 452 |
+
"grad_norm": 8.2695951461792,
|
| 453 |
+
"learning_rate": 1.4520547945205482e-05,
|
| 454 |
+
"loss": 0.126,
|
| 455 |
+
"step": 28000
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 6.971624266144814,
|
| 459 |
+
"grad_norm": 7.226652145385742,
|
| 460 |
+
"learning_rate": 1.4422700587084152e-05,
|
| 461 |
+
"loss": 0.1238,
|
| 462 |
+
"step": 28500
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 7.0,
|
| 466 |
+
"eval_accuracy": 0.9069863437790238,
|
| 467 |
+
"eval_loss": 0.3151151239871979,
|
| 468 |
+
"eval_runtime": 77.3038,
|
| 469 |
+
"eval_samples_per_second": 376.062,
|
| 470 |
+
"eval_steps_per_second": 11.759,
|
| 471 |
+
"step": 28616
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 7.0939334637964775,
|
| 475 |
+
"grad_norm": 6.771386623382568,
|
| 476 |
+
"learning_rate": 1.4324853228962818e-05,
|
| 477 |
+
"loss": 0.1014,
|
| 478 |
+
"step": 29000
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 7.216242661448141,
|
| 482 |
+
"grad_norm": 7.908430099487305,
|
| 483 |
+
"learning_rate": 1.4227005870841488e-05,
|
| 484 |
+
"loss": 0.0917,
|
| 485 |
+
"step": 29500
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 7.3385518590998045,
|
| 489 |
+
"grad_norm": 11.553496360778809,
|
| 490 |
+
"learning_rate": 1.4129158512720158e-05,
|
| 491 |
+
"loss": 0.0989,
|
| 492 |
+
"step": 30000
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 7.460861056751468,
|
| 496 |
+
"grad_norm": 5.48296594619751,
|
| 497 |
+
"learning_rate": 1.4031311154598828e-05,
|
| 498 |
+
"loss": 0.0945,
|
| 499 |
+
"step": 30500
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 7.583170254403131,
|
| 503 |
+
"grad_norm": 4.382979869842529,
|
| 504 |
+
"learning_rate": 1.3933463796477496e-05,
|
| 505 |
+
"loss": 0.0904,
|
| 506 |
+
"step": 31000
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 7.705479452054795,
|
| 510 |
+
"grad_norm": 4.479061126708984,
|
| 511 |
+
"learning_rate": 1.3835616438356164e-05,
|
| 512 |
+
"loss": 0.1003,
|
| 513 |
+
"step": 31500
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 7.8277886497064575,
|
| 517 |
+
"grad_norm": 2.2252697944641113,
|
| 518 |
+
"learning_rate": 1.3737769080234834e-05,
|
| 519 |
+
"loss": 0.0963,
|
| 520 |
+
"step": 32000
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 7.950097847358121,
|
| 524 |
+
"grad_norm": 5.149953842163086,
|
| 525 |
+
"learning_rate": 1.3639921722113504e-05,
|
| 526 |
+
"loss": 0.0969,
|
| 527 |
+
"step": 32500
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 8.0,
|
| 531 |
+
"eval_accuracy": 0.9127652987513329,
|
| 532 |
+
"eval_loss": 0.3146636486053467,
|
| 533 |
+
"eval_runtime": 77.5909,
|
| 534 |
+
"eval_samples_per_second": 374.67,
|
| 535 |
+
"eval_steps_per_second": 11.715,
|
| 536 |
+
"step": 32704
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 8.072407045009784,
|
| 540 |
+
"grad_norm": 10.789752960205078,
|
| 541 |
+
"learning_rate": 1.3542074363992173e-05,
|
| 542 |
+
"loss": 0.0788,
|
| 543 |
+
"step": 33000
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 8.194716242661448,
|
| 547 |
+
"grad_norm": 5.515634536743164,
|
| 548 |
+
"learning_rate": 1.3444227005870843e-05,
|
| 549 |
+
"loss": 0.0717,
|
| 550 |
+
"step": 33500
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 8.317025440313111,
|
| 554 |
+
"grad_norm": 10.454425811767578,
|
| 555 |
+
"learning_rate": 1.3346379647749513e-05,
|
| 556 |
+
"loss": 0.0719,
|
| 557 |
+
"step": 34000
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 8.439334637964775,
|
| 561 |
+
"grad_norm": 7.216288089752197,
|
| 562 |
+
"learning_rate": 1.3248532289628183e-05,
|
| 563 |
+
"loss": 0.0764,
|
| 564 |
+
"step": 34500
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 8.561643835616438,
|
| 568 |
+
"grad_norm": 6.058413028717041,
|
| 569 |
+
"learning_rate": 1.3150684931506849e-05,
|
| 570 |
+
"loss": 0.0775,
|
| 571 |
+
"step": 35000
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 8.683953033268102,
|
| 575 |
+
"grad_norm": 1.9321287870407104,
|
| 576 |
+
"learning_rate": 1.3052837573385519e-05,
|
| 577 |
+
"loss": 0.0762,
|
| 578 |
+
"step": 35500
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 8.806262230919765,
|
| 582 |
+
"grad_norm": 5.581308841705322,
|
| 583 |
+
"learning_rate": 1.2954990215264189e-05,
|
| 584 |
+
"loss": 0.0763,
|
| 585 |
+
"step": 36000
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 8.928571428571429,
|
| 589 |
+
"grad_norm": 8.296586036682129,
|
| 590 |
+
"learning_rate": 1.2857142857142859e-05,
|
| 591 |
+
"loss": 0.0826,
|
| 592 |
+
"step": 36500
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 9.0,
|
| 596 |
+
"eval_accuracy": 0.9133156754153624,
|
| 597 |
+
"eval_loss": 0.33166706562042236,
|
| 598 |
+
"eval_runtime": 77.4584,
|
| 599 |
+
"eval_samples_per_second": 375.311,
|
| 600 |
+
"eval_steps_per_second": 11.735,
|
| 601 |
+
"step": 36792
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 9.050880626223092,
|
| 605 |
+
"grad_norm": 6.40161657333374,
|
| 606 |
+
"learning_rate": 1.2759295499021527e-05,
|
| 607 |
+
"loss": 0.0689,
|
| 608 |
+
"step": 37000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 9.173189823874756,
|
| 612 |
+
"grad_norm": 9.357067108154297,
|
| 613 |
+
"learning_rate": 1.2661448140900197e-05,
|
| 614 |
+
"loss": 0.0581,
|
| 615 |
+
"step": 37500
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 9.29549902152642,
|
| 619 |
+
"grad_norm": 4.852538108825684,
|
| 620 |
+
"learning_rate": 1.2563600782778867e-05,
|
| 621 |
+
"loss": 0.0582,
|
| 622 |
+
"step": 38000
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 9.417808219178083,
|
| 626 |
+
"grad_norm": 6.204359531402588,
|
| 627 |
+
"learning_rate": 1.2465753424657537e-05,
|
| 628 |
+
"loss": 0.06,
|
| 629 |
+
"step": 38500
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 9.540117416829746,
|
| 633 |
+
"grad_norm": 11.658947944641113,
|
| 634 |
+
"learning_rate": 1.2367906066536204e-05,
|
| 635 |
+
"loss": 0.0626,
|
| 636 |
+
"step": 39000
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 9.66242661448141,
|
| 640 |
+
"grad_norm": 12.047215461730957,
|
| 641 |
+
"learning_rate": 1.2270058708414874e-05,
|
| 642 |
+
"loss": 0.0626,
|
| 643 |
+
"step": 39500
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 9.784735812133073,
|
| 647 |
+
"grad_norm": 11.078405380249023,
|
| 648 |
+
"learning_rate": 1.2172211350293543e-05,
|
| 649 |
+
"loss": 0.0655,
|
| 650 |
+
"step": 40000
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 9.907045009784735,
|
| 654 |
+
"grad_norm": 3.5147736072540283,
|
| 655 |
+
"learning_rate": 1.2074363992172213e-05,
|
| 656 |
+
"loss": 0.0663,
|
| 657 |
+
"step": 40500
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 10.0,
|
| 661 |
+
"eval_accuracy": 0.9173747033125795,
|
| 662 |
+
"eval_loss": 0.34417206048965454,
|
| 663 |
+
"eval_runtime": 77.3814,
|
| 664 |
+
"eval_samples_per_second": 375.685,
|
| 665 |
+
"eval_steps_per_second": 11.747,
|
| 666 |
+
"step": 40880
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 10.029354207436398,
|
| 670 |
+
"grad_norm": 2.839688539505005,
|
| 671 |
+
"learning_rate": 1.1976516634050882e-05,
|
| 672 |
+
"loss": 0.0609,
|
| 673 |
+
"step": 41000
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 10.151663405088062,
|
| 677 |
+
"grad_norm": 10.794103622436523,
|
| 678 |
+
"learning_rate": 1.187866927592955e-05,
|
| 679 |
+
"loss": 0.0467,
|
| 680 |
+
"step": 41500
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 10.273972602739725,
|
| 684 |
+
"grad_norm": 3.0600426197052,
|
| 685 |
+
"learning_rate": 1.178082191780822e-05,
|
| 686 |
+
"loss": 0.0464,
|
| 687 |
+
"step": 42000
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 10.396281800391389,
|
| 691 |
+
"grad_norm": 7.135795593261719,
|
| 692 |
+
"learning_rate": 1.168297455968689e-05,
|
| 693 |
+
"loss": 0.051,
|
| 694 |
+
"step": 42500
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 10.518590998043052,
|
| 698 |
+
"grad_norm": 5.585446834564209,
|
| 699 |
+
"learning_rate": 1.1585127201565558e-05,
|
| 700 |
+
"loss": 0.0531,
|
| 701 |
+
"step": 43000
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 10.640900195694716,
|
| 705 |
+
"grad_norm": 0.05372029170393944,
|
| 706 |
+
"learning_rate": 1.1487279843444228e-05,
|
| 707 |
+
"loss": 0.0504,
|
| 708 |
+
"step": 43500
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 10.76320939334638,
|
| 712 |
+
"grad_norm": 12.072264671325684,
|
| 713 |
+
"learning_rate": 1.1389432485322898e-05,
|
| 714 |
+
"loss": 0.0511,
|
| 715 |
+
"step": 44000
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 10.885518590998043,
|
| 719 |
+
"grad_norm": 5.787899971008301,
|
| 720 |
+
"learning_rate": 1.1291585127201568e-05,
|
| 721 |
+
"loss": 0.0519,
|
| 722 |
+
"step": 44500
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 11.0,
|
| 726 |
+
"eval_accuracy": 0.9171339135220666,
|
| 727 |
+
"eval_loss": 0.3852131962776184,
|
| 728 |
+
"eval_runtime": 77.3953,
|
| 729 |
+
"eval_samples_per_second": 375.617,
|
| 730 |
+
"eval_steps_per_second": 11.745,
|
| 731 |
+
"step": 44968
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 11.007827788649706,
|
| 735 |
+
"grad_norm": 8.351863861083984,
|
| 736 |
+
"learning_rate": 1.1193737769080235e-05,
|
| 737 |
+
"loss": 0.0508,
|
| 738 |
+
"step": 45000
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 11.13013698630137,
|
| 742 |
+
"grad_norm": 6.567861080169678,
|
| 743 |
+
"learning_rate": 1.1095890410958904e-05,
|
| 744 |
+
"loss": 0.0403,
|
| 745 |
+
"step": 45500
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 11.252446183953033,
|
| 749 |
+
"grad_norm": 5.368165016174316,
|
| 750 |
+
"learning_rate": 1.0998043052837574e-05,
|
| 751 |
+
"loss": 0.0419,
|
| 752 |
+
"step": 46000
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 11.374755381604697,
|
| 756 |
+
"grad_norm": 11.540104866027832,
|
| 757 |
+
"learning_rate": 1.0900195694716244e-05,
|
| 758 |
+
"loss": 0.0466,
|
| 759 |
+
"step": 46500
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 11.49706457925636,
|
| 763 |
+
"grad_norm": 2.5703656673431396,
|
| 764 |
+
"learning_rate": 1.0802348336594913e-05,
|
| 765 |
+
"loss": 0.0422,
|
| 766 |
+
"step": 47000
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 11.619373776908024,
|
| 770 |
+
"grad_norm": 10.161659240722656,
|
| 771 |
+
"learning_rate": 1.0704500978473583e-05,
|
| 772 |
+
"loss": 0.0492,
|
| 773 |
+
"step": 47500
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 11.741682974559687,
|
| 777 |
+
"grad_norm": 9.083320617675781,
|
| 778 |
+
"learning_rate": 1.0606653620352253e-05,
|
| 779 |
+
"loss": 0.0444,
|
| 780 |
+
"step": 48000
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 11.86399217221135,
|
| 784 |
+
"grad_norm": 7.005116939544678,
|
| 785 |
+
"learning_rate": 1.0508806262230922e-05,
|
| 786 |
+
"loss": 0.0431,
|
| 787 |
+
"step": 48500
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 11.986301369863014,
|
| 791 |
+
"grad_norm": 5.291450023651123,
|
| 792 |
+
"learning_rate": 1.0410958904109589e-05,
|
| 793 |
+
"loss": 0.045,
|
| 794 |
+
"step": 49000
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 12.0,
|
| 798 |
+
"eval_accuracy": 0.9189226376801624,
|
| 799 |
+
"eval_loss": 0.3933347463607788,
|
| 800 |
+
"eval_runtime": 77.3468,
|
| 801 |
+
"eval_samples_per_second": 375.852,
|
| 802 |
+
"eval_steps_per_second": 11.752,
|
| 803 |
+
"step": 49056
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 12.108610567514678,
|
| 807 |
+
"grad_norm": 3.0380940437316895,
|
| 808 |
+
"learning_rate": 1.0313111545988259e-05,
|
| 809 |
+
"loss": 0.0338,
|
| 810 |
+
"step": 49500
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 12.230919765166341,
|
| 814 |
+
"grad_norm": 0.6975650787353516,
|
| 815 |
+
"learning_rate": 1.0215264187866929e-05,
|
| 816 |
+
"loss": 0.0343,
|
| 817 |
+
"step": 50000
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 12.353228962818005,
|
| 821 |
+
"grad_norm": 0.18672548234462738,
|
| 822 |
+
"learning_rate": 1.0117416829745599e-05,
|
| 823 |
+
"loss": 0.0349,
|
| 824 |
+
"step": 50500
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 12.475538160469668,
|
| 828 |
+
"grad_norm": 6.868373394012451,
|
| 829 |
+
"learning_rate": 1.0019569471624267e-05,
|
| 830 |
+
"loss": 0.0343,
|
| 831 |
+
"step": 51000
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 12.597847358121331,
|
| 835 |
+
"grad_norm": 1.4025439023971558,
|
| 836 |
+
"learning_rate": 9.921722113502935e-06,
|
| 837 |
+
"loss": 0.0396,
|
| 838 |
+
"step": 51500
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 12.720156555772995,
|
| 842 |
+
"grad_norm": 4.942438125610352,
|
| 843 |
+
"learning_rate": 9.823874755381605e-06,
|
| 844 |
+
"loss": 0.0381,
|
| 845 |
+
"step": 52000
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 12.842465753424658,
|
| 849 |
+
"grad_norm": 4.044002056121826,
|
| 850 |
+
"learning_rate": 9.726027397260275e-06,
|
| 851 |
+
"loss": 0.0382,
|
| 852 |
+
"step": 52500
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 12.96477495107632,
|
| 856 |
+
"grad_norm": 0.6736146211624146,
|
| 857 |
+
"learning_rate": 9.628180039138944e-06,
|
| 858 |
+
"loss": 0.0368,
|
| 859 |
+
"step": 53000
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 13.0,
|
| 863 |
+
"eval_accuracy": 0.9228096728698704,
|
| 864 |
+
"eval_loss": 0.38157734274864197,
|
| 865 |
+
"eval_runtime": 77.3834,
|
| 866 |
+
"eval_samples_per_second": 375.675,
|
| 867 |
+
"eval_steps_per_second": 11.747,
|
| 868 |
+
"step": 53144
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 13.087084148727984,
|
| 872 |
+
"grad_norm": 1.1303954124450684,
|
| 873 |
+
"learning_rate": 9.530332681017614e-06,
|
| 874 |
+
"loss": 0.0306,
|
| 875 |
+
"step": 53500
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 13.209393346379647,
|
| 879 |
+
"grad_norm": 1.899757742881775,
|
| 880 |
+
"learning_rate": 9.432485322896282e-06,
|
| 881 |
+
"loss": 0.0302,
|
| 882 |
+
"step": 54000
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 13.33170254403131,
|
| 886 |
+
"grad_norm": 10.160910606384277,
|
| 887 |
+
"learning_rate": 9.334637964774952e-06,
|
| 888 |
+
"loss": 0.0284,
|
| 889 |
+
"step": 54500
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 13.454011741682974,
|
| 893 |
+
"grad_norm": 6.715288162231445,
|
| 894 |
+
"learning_rate": 9.23679060665362e-06,
|
| 895 |
+
"loss": 0.0351,
|
| 896 |
+
"step": 55000
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 13.576320939334638,
|
| 900 |
+
"grad_norm": 9.383013725280762,
|
| 901 |
+
"learning_rate": 9.13894324853229e-06,
|
| 902 |
+
"loss": 0.0342,
|
| 903 |
+
"step": 55500
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 13.698630136986301,
|
| 907 |
+
"grad_norm": 0.5024349093437195,
|
| 908 |
+
"learning_rate": 9.04109589041096e-06,
|
| 909 |
+
"loss": 0.0358,
|
| 910 |
+
"step": 56000
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 13.820939334637965,
|
| 914 |
+
"grad_norm": 6.495013236999512,
|
| 915 |
+
"learning_rate": 8.943248532289628e-06,
|
| 916 |
+
"loss": 0.0361,
|
| 917 |
+
"step": 56500
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 13.943248532289628,
|
| 921 |
+
"grad_norm": 3.360276699066162,
|
| 922 |
+
"learning_rate": 8.845401174168298e-06,
|
| 923 |
+
"loss": 0.0332,
|
| 924 |
+
"step": 57000
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 14.0,
|
| 928 |
+
"eval_accuracy": 0.9229816655773795,
|
| 929 |
+
"eval_loss": 0.4262143075466156,
|
| 930 |
+
"eval_runtime": 77.1924,
|
| 931 |
+
"eval_samples_per_second": 376.604,
|
| 932 |
+
"eval_steps_per_second": 11.776,
|
| 933 |
+
"step": 57232
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 14.065557729941291,
|
| 937 |
+
"grad_norm": 2.5697081089019775,
|
| 938 |
+
"learning_rate": 8.747553816046968e-06,
|
| 939 |
+
"loss": 0.0271,
|
| 940 |
+
"step": 57500
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 14.187866927592955,
|
| 944 |
+
"grad_norm": 2.5552070140838623,
|
| 945 |
+
"learning_rate": 8.649706457925636e-06,
|
| 946 |
+
"loss": 0.0229,
|
| 947 |
+
"step": 58000
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 14.310176125244618,
|
| 951 |
+
"grad_norm": 11.48392105102539,
|
| 952 |
+
"learning_rate": 8.551859099804306e-06,
|
| 953 |
+
"loss": 0.0275,
|
| 954 |
+
"step": 58500
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 14.432485322896282,
|
| 958 |
+
"grad_norm": 4.35728645324707,
|
| 959 |
+
"learning_rate": 8.454011741682975e-06,
|
| 960 |
+
"loss": 0.026,
|
| 961 |
+
"step": 59000
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 14.554794520547945,
|
| 965 |
+
"grad_norm": 18.579906463623047,
|
| 966 |
+
"learning_rate": 8.356164383561644e-06,
|
| 967 |
+
"loss": 0.0259,
|
| 968 |
+
"step": 59500
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 14.677103718199609,
|
| 972 |
+
"grad_norm": 7.566853046417236,
|
| 973 |
+
"learning_rate": 8.258317025440313e-06,
|
| 974 |
+
"loss": 0.0281,
|
| 975 |
+
"step": 60000
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 14.799412915851272,
|
| 979 |
+
"grad_norm": 0.4659290611743927,
|
| 980 |
+
"learning_rate": 8.160469667318983e-06,
|
| 981 |
+
"loss": 0.0283,
|
| 982 |
+
"step": 60500
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 14.921722113502936,
|
| 986 |
+
"grad_norm": 5.014148235321045,
|
| 987 |
+
"learning_rate": 8.062622309197653e-06,
|
| 988 |
+
"loss": 0.0289,
|
| 989 |
+
"step": 61000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 15.0,
|
| 993 |
+
"eval_accuracy": 0.9236008393244126,
|
| 994 |
+
"eval_loss": 0.43280109763145447,
|
| 995 |
+
"eval_runtime": 77.9069,
|
| 996 |
+
"eval_samples_per_second": 373.151,
|
| 997 |
+
"eval_steps_per_second": 11.668,
|
| 998 |
+
"step": 61320
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 15.0440313111546,
|
| 1002 |
+
"grad_norm": 4.811933994293213,
|
| 1003 |
+
"learning_rate": 7.964774951076321e-06,
|
| 1004 |
+
"loss": 0.027,
|
| 1005 |
+
"step": 61500
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 15.166340508806263,
|
| 1009 |
+
"grad_norm": 9.811365127563477,
|
| 1010 |
+
"learning_rate": 7.86692759295499e-06,
|
| 1011 |
+
"loss": 0.0214,
|
| 1012 |
+
"step": 62000
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 15.288649706457926,
|
| 1016 |
+
"grad_norm": 1.3940095901489258,
|
| 1017 |
+
"learning_rate": 7.76908023483366e-06,
|
| 1018 |
+
"loss": 0.0219,
|
| 1019 |
+
"step": 62500
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 15.41095890410959,
|
| 1023 |
+
"grad_norm": 0.06226726993918419,
|
| 1024 |
+
"learning_rate": 7.671232876712329e-06,
|
| 1025 |
+
"loss": 0.0226,
|
| 1026 |
+
"step": 63000
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 15.533268101761253,
|
| 1030 |
+
"grad_norm": 8.294939994812012,
|
| 1031 |
+
"learning_rate": 7.573385518590999e-06,
|
| 1032 |
+
"loss": 0.0251,
|
| 1033 |
+
"step": 63500
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 15.655577299412915,
|
| 1037 |
+
"grad_norm": 1.8952441215515137,
|
| 1038 |
+
"learning_rate": 7.475538160469667e-06,
|
| 1039 |
+
"loss": 0.0232,
|
| 1040 |
+
"step": 64000
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 15.777886497064578,
|
| 1044 |
+
"grad_norm": 5.1830315589904785,
|
| 1045 |
+
"learning_rate": 7.377690802348337e-06,
|
| 1046 |
+
"loss": 0.0254,
|
| 1047 |
+
"step": 64500
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 15.900195694716242,
|
| 1051 |
+
"grad_norm": 1.0473072528839111,
|
| 1052 |
+
"learning_rate": 7.279843444227006e-06,
|
| 1053 |
+
"loss": 0.0248,
|
| 1054 |
+
"step": 65000
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 16.0,
|
| 1058 |
+
"eval_accuracy": 0.9260431357710434,
|
| 1059 |
+
"eval_loss": 0.4366961419582367,
|
| 1060 |
+
"eval_runtime": 77.3755,
|
| 1061 |
+
"eval_samples_per_second": 375.713,
|
| 1062 |
+
"eval_steps_per_second": 11.748,
|
| 1063 |
+
"step": 65408
|
| 1064 |
+
},
|
| 1065 |
+
{
|
| 1066 |
+
"epoch": 16.022504892367905,
|
| 1067 |
+
"grad_norm": 0.5687731504440308,
|
| 1068 |
+
"learning_rate": 7.181996086105676e-06,
|
| 1069 |
+
"loss": 0.0238,
|
| 1070 |
+
"step": 65500
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 16.14481409001957,
|
| 1074 |
+
"grad_norm": 1.2872580289840698,
|
| 1075 |
+
"learning_rate": 7.0841487279843445e-06,
|
| 1076 |
+
"loss": 0.0194,
|
| 1077 |
+
"step": 66000
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 16.267123287671232,
|
| 1081 |
+
"grad_norm": 5.059693813323975,
|
| 1082 |
+
"learning_rate": 6.9863013698630145e-06,
|
| 1083 |
+
"loss": 0.0199,
|
| 1084 |
+
"step": 66500
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 16.389432485322896,
|
| 1088 |
+
"grad_norm": 6.738853931427002,
|
| 1089 |
+
"learning_rate": 6.8884540117416836e-06,
|
| 1090 |
+
"loss": 0.0205,
|
| 1091 |
+
"step": 67000
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 16.51174168297456,
|
| 1095 |
+
"grad_norm": 11.091848373413086,
|
| 1096 |
+
"learning_rate": 6.790606653620353e-06,
|
| 1097 |
+
"loss": 0.0215,
|
| 1098 |
+
"step": 67500
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 16.634050880626223,
|
| 1102 |
+
"grad_norm": 0.19167885184288025,
|
| 1103 |
+
"learning_rate": 6.692759295499022e-06,
|
| 1104 |
+
"loss": 0.022,
|
| 1105 |
+
"step": 68000
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 16.756360078277886,
|
| 1109 |
+
"grad_norm": 5.966000080108643,
|
| 1110 |
+
"learning_rate": 6.594911937377692e-06,
|
| 1111 |
+
"loss": 0.0203,
|
| 1112 |
+
"step": 68500
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 16.87866927592955,
|
| 1116 |
+
"grad_norm": 3.5066215991973877,
|
| 1117 |
+
"learning_rate": 6.49706457925636e-06,
|
| 1118 |
+
"loss": 0.0187,
|
| 1119 |
+
"step": 69000
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 17.0,
|
| 1123 |
+
"eval_accuracy": 0.9272470847236077,
|
| 1124 |
+
"eval_loss": 0.44777625799179077,
|
| 1125 |
+
"eval_runtime": 77.1492,
|
| 1126 |
+
"eval_samples_per_second": 376.815,
|
| 1127 |
+
"eval_steps_per_second": 11.782,
|
| 1128 |
+
"step": 69496
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 17.000978473581213,
|
| 1132 |
+
"grad_norm": 0.6858524680137634,
|
| 1133 |
+
"learning_rate": 6.39921722113503e-06,
|
| 1134 |
+
"loss": 0.0204,
|
| 1135 |
+
"step": 69500
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 17.123287671232877,
|
| 1139 |
+
"grad_norm": 9.703988075256348,
|
| 1140 |
+
"learning_rate": 6.301369863013699e-06,
|
| 1141 |
+
"loss": 0.0161,
|
| 1142 |
+
"step": 70000
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 17.24559686888454,
|
| 1146 |
+
"grad_norm": 4.51020622253418,
|
| 1147 |
+
"learning_rate": 6.203522504892369e-06,
|
| 1148 |
+
"loss": 0.016,
|
| 1149 |
+
"step": 70500
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 17.367906066536204,
|
| 1153 |
+
"grad_norm": 0.09826533496379852,
|
| 1154 |
+
"learning_rate": 6.105675146771037e-06,
|
| 1155 |
+
"loss": 0.0177,
|
| 1156 |
+
"step": 71000
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 17.490215264187867,
|
| 1160 |
+
"grad_norm": 11.620870590209961,
|
| 1161 |
+
"learning_rate": 6.007827788649707e-06,
|
| 1162 |
+
"loss": 0.02,
|
| 1163 |
+
"step": 71500
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 17.61252446183953,
|
| 1167 |
+
"grad_norm": 8.661576271057129,
|
| 1168 |
+
"learning_rate": 5.909980430528376e-06,
|
| 1169 |
+
"loss": 0.017,
|
| 1170 |
+
"step": 72000
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 17.734833659491194,
|
| 1174 |
+
"grad_norm": 0.7512950897216797,
|
| 1175 |
+
"learning_rate": 5.812133072407045e-06,
|
| 1176 |
+
"loss": 0.0186,
|
| 1177 |
+
"step": 72500
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 17.857142857142858,
|
| 1181 |
+
"grad_norm": 0.44471457600593567,
|
| 1182 |
+
"learning_rate": 5.7142857142857145e-06,
|
| 1183 |
+
"loss": 0.0228,
|
| 1184 |
+
"step": 73000
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 17.97945205479452,
|
| 1188 |
+
"grad_norm": 1.0926498174667358,
|
| 1189 |
+
"learning_rate": 5.6164383561643845e-06,
|
| 1190 |
+
"loss": 0.0178,
|
| 1191 |
+
"step": 73500
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 18.0,
|
| 1195 |
+
"eval_accuracy": 0.9264559182690654,
|
| 1196 |
+
"eval_loss": 0.46360349655151367,
|
| 1197 |
+
"eval_runtime": 77.3263,
|
| 1198 |
+
"eval_samples_per_second": 375.952,
|
| 1199 |
+
"eval_steps_per_second": 11.755,
|
| 1200 |
+
"step": 73584
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 18.101761252446185,
|
| 1204 |
+
"grad_norm": 8.134523391723633,
|
| 1205 |
+
"learning_rate": 5.518590998043053e-06,
|
| 1206 |
+
"loss": 0.0153,
|
| 1207 |
+
"step": 74000
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 18.224070450097848,
|
| 1211 |
+
"grad_norm": 0.08053633570671082,
|
| 1212 |
+
"learning_rate": 5.420743639921723e-06,
|
| 1213 |
+
"loss": 0.016,
|
| 1214 |
+
"step": 74500
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 18.34637964774951,
|
| 1218 |
+
"grad_norm": 0.44435006380081177,
|
| 1219 |
+
"learning_rate": 5.322896281800392e-06,
|
| 1220 |
+
"loss": 0.0146,
|
| 1221 |
+
"step": 75000
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 18.468688845401175,
|
| 1225 |
+
"grad_norm": 2.0229480266571045,
|
| 1226 |
+
"learning_rate": 5.225048923679062e-06,
|
| 1227 |
+
"loss": 0.017,
|
| 1228 |
+
"step": 75500
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 18.59099804305284,
|
| 1232 |
+
"grad_norm": 4.502400875091553,
|
| 1233 |
+
"learning_rate": 5.12720156555773e-06,
|
| 1234 |
+
"loss": 0.017,
|
| 1235 |
+
"step": 76000
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 18.713307240704502,
|
| 1239 |
+
"grad_norm": 0.14978571236133575,
|
| 1240 |
+
"learning_rate": 5.0293542074364e-06,
|
| 1241 |
+
"loss": 0.0154,
|
| 1242 |
+
"step": 76500
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 18.835616438356166,
|
| 1246 |
+
"grad_norm": 9.524068832397461,
|
| 1247 |
+
"learning_rate": 4.931506849315069e-06,
|
| 1248 |
+
"loss": 0.0165,
|
| 1249 |
+
"step": 77000
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 18.95792563600783,
|
| 1253 |
+
"grad_norm": 0.9707121253013611,
|
| 1254 |
+
"learning_rate": 4.833659491193738e-06,
|
| 1255 |
+
"loss": 0.0154,
|
| 1256 |
+
"step": 77500
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 19.0,
|
| 1260 |
+
"eval_accuracy": 0.9266623095180765,
|
| 1261 |
+
"eval_loss": 0.47839030623435974,
|
| 1262 |
+
"eval_runtime": 77.6682,
|
| 1263 |
+
"eval_samples_per_second": 374.297,
|
| 1264 |
+
"eval_steps_per_second": 11.704,
|
| 1265 |
+
"step": 77672
|
| 1266 |
+
},
|
| 1267 |
+
{
|
| 1268 |
+
"epoch": 19.080234833659492,
|
| 1269 |
+
"grad_norm": 11.346772193908691,
|
| 1270 |
+
"learning_rate": 4.735812133072407e-06,
|
| 1271 |
+
"loss": 0.0136,
|
| 1272 |
+
"step": 78000
|
| 1273 |
+
},
|
| 1274 |
+
{
|
| 1275 |
+
"epoch": 19.202544031311156,
|
| 1276 |
+
"grad_norm": 0.7873362302780151,
|
| 1277 |
+
"learning_rate": 4.637964774951076e-06,
|
| 1278 |
+
"loss": 0.0131,
|
| 1279 |
+
"step": 78500
|
| 1280 |
+
},
|
| 1281 |
+
{
|
| 1282 |
+
"epoch": 19.32485322896282,
|
| 1283 |
+
"grad_norm": 1.0734236240386963,
|
| 1284 |
+
"learning_rate": 4.5401174168297455e-06,
|
| 1285 |
+
"loss": 0.0156,
|
| 1286 |
+
"step": 79000
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 19.447162426614483,
|
| 1290 |
+
"grad_norm": 0.031163902953267097,
|
| 1291 |
+
"learning_rate": 4.442270058708415e-06,
|
| 1292 |
+
"loss": 0.0148,
|
| 1293 |
+
"step": 79500
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 19.569471624266146,
|
| 1297 |
+
"grad_norm": 3.0132200717926025,
|
| 1298 |
+
"learning_rate": 4.3444227005870845e-06,
|
| 1299 |
+
"loss": 0.0142,
|
| 1300 |
+
"step": 80000
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 19.69178082191781,
|
| 1304 |
+
"grad_norm": 5.072915554046631,
|
| 1305 |
+
"learning_rate": 4.246575342465754e-06,
|
| 1306 |
+
"loss": 0.0133,
|
| 1307 |
+
"step": 80500
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 19.81409001956947,
|
| 1311 |
+
"grad_norm": 0.12334894388914108,
|
| 1312 |
+
"learning_rate": 4.148727984344423e-06,
|
| 1313 |
+
"loss": 0.0153,
|
| 1314 |
+
"step": 81000
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 19.936399217221137,
|
| 1318 |
+
"grad_norm": 3.8586535453796387,
|
| 1319 |
+
"learning_rate": 4.050880626223092e-06,
|
| 1320 |
+
"loss": 0.0138,
|
| 1321 |
+
"step": 81500
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 20.0,
|
| 1325 |
+
"eval_accuracy": 0.9280726497196519,
|
| 1326 |
+
"eval_loss": 0.4880678355693817,
|
| 1327 |
+
"eval_runtime": 77.1562,
|
| 1328 |
+
"eval_samples_per_second": 376.781,
|
| 1329 |
+
"eval_steps_per_second": 11.781,
|
| 1330 |
+
"step": 81760
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 20.058708414872797,
|
| 1334 |
+
"grad_norm": 5.422935485839844,
|
| 1335 |
+
"learning_rate": 3.953033268101762e-06,
|
| 1336 |
+
"loss": 0.0123,
|
| 1337 |
+
"step": 82000
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 20.18101761252446,
|
| 1341 |
+
"grad_norm": 0.02908429130911827,
|
| 1342 |
+
"learning_rate": 3.855185909980431e-06,
|
| 1343 |
+
"loss": 0.0127,
|
| 1344 |
+
"step": 82500
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 20.303326810176124,
|
| 1348 |
+
"grad_norm": 0.9683161973953247,
|
| 1349 |
+
"learning_rate": 3.7573385518591e-06,
|
| 1350 |
+
"loss": 0.0106,
|
| 1351 |
+
"step": 83000
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 20.425636007827787,
|
| 1355 |
+
"grad_norm": 2.011960029602051,
|
| 1356 |
+
"learning_rate": 3.659491193737769e-06,
|
| 1357 |
+
"loss": 0.0129,
|
| 1358 |
+
"step": 83500
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 20.54794520547945,
|
| 1362 |
+
"grad_norm": 0.011290821246802807,
|
| 1363 |
+
"learning_rate": 3.5616438356164386e-06,
|
| 1364 |
+
"loss": 0.0098,
|
| 1365 |
+
"step": 84000
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 20.670254403131114,
|
| 1369 |
+
"grad_norm": 3.6643381118774414,
|
| 1370 |
+
"learning_rate": 3.4637964774951077e-06,
|
| 1371 |
+
"loss": 0.0123,
|
| 1372 |
+
"step": 84500
|
| 1373 |
+
},
|
| 1374 |
+
{
|
| 1375 |
+
"epoch": 20.792563600782778,
|
| 1376 |
+
"grad_norm": 0.28946495056152344,
|
| 1377 |
+
"learning_rate": 3.365949119373777e-06,
|
| 1378 |
+
"loss": 0.0119,
|
| 1379 |
+
"step": 85000
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 20.91487279843444,
|
| 1383 |
+
"grad_norm": 0.8526332974433899,
|
| 1384 |
+
"learning_rate": 3.2681017612524463e-06,
|
| 1385 |
+
"loss": 0.0127,
|
| 1386 |
+
"step": 85500
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 21.0,
|
| 1390 |
+
"eval_accuracy": 0.9281758453441574,
|
| 1391 |
+
"eval_loss": 0.49737828969955444,
|
| 1392 |
+
"eval_runtime": 77.6066,
|
| 1393 |
+
"eval_samples_per_second": 374.594,
|
| 1394 |
+
"eval_steps_per_second": 11.713,
|
| 1395 |
+
"step": 85848
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 21.037181996086105,
|
| 1399 |
+
"grad_norm": 0.038626112043857574,
|
| 1400 |
+
"learning_rate": 3.1702544031311154e-06,
|
| 1401 |
+
"loss": 0.0133,
|
| 1402 |
+
"step": 86000
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 21.159491193737768,
|
| 1406 |
+
"grad_norm": 1.5674916505813599,
|
| 1407 |
+
"learning_rate": 3.072407045009785e-06,
|
| 1408 |
+
"loss": 0.0079,
|
| 1409 |
+
"step": 86500
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 21.28180039138943,
|
| 1413 |
+
"grad_norm": 0.023863431066274643,
|
| 1414 |
+
"learning_rate": 2.974559686888454e-06,
|
| 1415 |
+
"loss": 0.0089,
|
| 1416 |
+
"step": 87000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 21.404109589041095,
|
| 1420 |
+
"grad_norm": 0.027295144274830818,
|
| 1421 |
+
"learning_rate": 2.876712328767123e-06,
|
| 1422 |
+
"loss": 0.0099,
|
| 1423 |
+
"step": 87500
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 21.52641878669276,
|
| 1427 |
+
"grad_norm": 1.8621610403060913,
|
| 1428 |
+
"learning_rate": 2.7788649706457927e-06,
|
| 1429 |
+
"loss": 0.01,
|
| 1430 |
+
"step": 88000
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 21.648727984344422,
|
| 1434 |
+
"grad_norm": 0.09760987758636475,
|
| 1435 |
+
"learning_rate": 2.681017612524462e-06,
|
| 1436 |
+
"loss": 0.0115,
|
| 1437 |
+
"step": 88500
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 21.771037181996086,
|
| 1441 |
+
"grad_norm": 0.0476505346596241,
|
| 1442 |
+
"learning_rate": 2.5831702544031313e-06,
|
| 1443 |
+
"loss": 0.0096,
|
| 1444 |
+
"step": 89000
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 21.89334637964775,
|
| 1448 |
+
"grad_norm": 0.08127936720848083,
|
| 1449 |
+
"learning_rate": 2.4853228962818004e-06,
|
| 1450 |
+
"loss": 0.0116,
|
| 1451 |
+
"step": 89500
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 22.0,
|
| 1455 |
+
"eval_accuracy": 0.9293797942967218,
|
| 1456 |
+
"eval_loss": 0.4935356378555298,
|
| 1457 |
+
"eval_runtime": 79.4825,
|
| 1458 |
+
"eval_samples_per_second": 365.754,
|
| 1459 |
+
"eval_steps_per_second": 11.436,
|
| 1460 |
+
"step": 89936
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 22.015655577299412,
|
| 1464 |
+
"grad_norm": 1.2286592721939087,
|
| 1465 |
+
"learning_rate": 2.3874755381604695e-06,
|
| 1466 |
+
"loss": 0.0086,
|
| 1467 |
+
"step": 90000
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 22.137964774951076,
|
| 1471 |
+
"grad_norm": 1.9946578741073608,
|
| 1472 |
+
"learning_rate": 2.289628180039139e-06,
|
| 1473 |
+
"loss": 0.008,
|
| 1474 |
+
"step": 90500
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 22.26027397260274,
|
| 1478 |
+
"grad_norm": 0.024408530443906784,
|
| 1479 |
+
"learning_rate": 2.191780821917808e-06,
|
| 1480 |
+
"loss": 0.0115,
|
| 1481 |
+
"step": 91000
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 22.382583170254403,
|
| 1485 |
+
"grad_norm": 0.011225187219679356,
|
| 1486 |
+
"learning_rate": 2.0939334637964777e-06,
|
| 1487 |
+
"loss": 0.0085,
|
| 1488 |
+
"step": 91500
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 22.504892367906066,
|
| 1492 |
+
"grad_norm": 1.2969368696212769,
|
| 1493 |
+
"learning_rate": 1.996086105675147e-06,
|
| 1494 |
+
"loss": 0.0089,
|
| 1495 |
+
"step": 92000
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 22.62720156555773,
|
| 1499 |
+
"grad_norm": 6.824276447296143,
|
| 1500 |
+
"learning_rate": 1.8982387475538161e-06,
|
| 1501 |
+
"loss": 0.0088,
|
| 1502 |
+
"step": 92500
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 22.749510763209393,
|
| 1506 |
+
"grad_norm": 0.052455250173807144,
|
| 1507 |
+
"learning_rate": 1.8003913894324854e-06,
|
| 1508 |
+
"loss": 0.0088,
|
| 1509 |
+
"step": 93000
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 22.871819960861057,
|
| 1513 |
+
"grad_norm": 0.04021551460027695,
|
| 1514 |
+
"learning_rate": 1.7025440313111545e-06,
|
| 1515 |
+
"loss": 0.009,
|
| 1516 |
+
"step": 93500
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 22.99412915851272,
|
| 1520 |
+
"grad_norm": 0.06902284175157547,
|
| 1521 |
+
"learning_rate": 1.6046966731898239e-06,
|
| 1522 |
+
"loss": 0.0106,
|
| 1523 |
+
"step": 94000
|
| 1524 |
+
},
|
| 1525 |
+
{
|
| 1526 |
+
"epoch": 23.0,
|
| 1527 |
+
"eval_accuracy": 0.93065254033229,
|
| 1528 |
+
"eval_loss": 0.4948473870754242,
|
| 1529 |
+
"eval_runtime": 79.6762,
|
| 1530 |
+
"eval_samples_per_second": 364.864,
|
| 1531 |
+
"eval_steps_per_second": 11.409,
|
| 1532 |
+
"step": 94024
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 23.116438356164384,
|
| 1536 |
+
"grad_norm": 0.5471883416175842,
|
| 1537 |
+
"learning_rate": 1.5068493150684932e-06,
|
| 1538 |
+
"loss": 0.0068,
|
| 1539 |
+
"step": 94500
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 23.238747553816047,
|
| 1543 |
+
"grad_norm": 0.16019482910633087,
|
| 1544 |
+
"learning_rate": 1.4090019569471625e-06,
|
| 1545 |
+
"loss": 0.0089,
|
| 1546 |
+
"step": 95000
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 23.36105675146771,
|
| 1550 |
+
"grad_norm": 6.4080963134765625,
|
| 1551 |
+
"learning_rate": 1.3111545988258318e-06,
|
| 1552 |
+
"loss": 0.0077,
|
| 1553 |
+
"step": 95500
|
| 1554 |
+
},
|
| 1555 |
+
{
|
| 1556 |
+
"epoch": 23.483365949119374,
|
| 1557 |
+
"grad_norm": 0.13792355358600616,
|
| 1558 |
+
"learning_rate": 1.213307240704501e-06,
|
| 1559 |
+
"loss": 0.0076,
|
| 1560 |
+
"step": 96000
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 23.605675146771038,
|
| 1564 |
+
"grad_norm": 2.563711404800415,
|
| 1565 |
+
"learning_rate": 1.1154598825831702e-06,
|
| 1566 |
+
"loss": 0.0075,
|
| 1567 |
+
"step": 96500
|
| 1568 |
+
},
|
| 1569 |
+
{
|
| 1570 |
+
"epoch": 23.7279843444227,
|
| 1571 |
+
"grad_norm": 0.04809940978884697,
|
| 1572 |
+
"learning_rate": 1.0176125244618395e-06,
|
| 1573 |
+
"loss": 0.0067,
|
| 1574 |
+
"step": 97000
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 23.850293542074365,
|
| 1578 |
+
"grad_norm": 0.057318106293678284,
|
| 1579 |
+
"learning_rate": 9.197651663405089e-07,
|
| 1580 |
+
"loss": 0.0062,
|
| 1581 |
+
"step": 97500
|
| 1582 |
+
},
|
| 1583 |
+
{
|
| 1584 |
+
"epoch": 23.972602739726028,
|
| 1585 |
+
"grad_norm": 4.935258865356445,
|
| 1586 |
+
"learning_rate": 8.219178082191781e-07,
|
| 1587 |
+
"loss": 0.0085,
|
| 1588 |
+
"step": 98000
|
| 1589 |
+
},
|
| 1590 |
+
{
|
| 1591 |
+
"epoch": 24.0,
|
| 1592 |
+
"eval_accuracy": 0.9311685184548175,
|
| 1593 |
+
"eval_loss": 0.49324169754981995,
|
| 1594 |
+
"eval_runtime": 79.011,
|
| 1595 |
+
"eval_samples_per_second": 367.936,
|
| 1596 |
+
"eval_steps_per_second": 11.505,
|
| 1597 |
+
"step": 98112
|
| 1598 |
+
},
|
| 1599 |
+
{
|
| 1600 |
+
"epoch": 24.09491193737769,
|
| 1601 |
+
"grad_norm": 6.268829345703125,
|
| 1602 |
+
"learning_rate": 7.240704500978474e-07,
|
| 1603 |
+
"loss": 0.0055,
|
| 1604 |
+
"step": 98500
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 24.217221135029355,
|
| 1608 |
+
"grad_norm": 0.00456130551174283,
|
| 1609 |
+
"learning_rate": 6.262230919765167e-07,
|
| 1610 |
+
"loss": 0.0054,
|
| 1611 |
+
"step": 99000
|
| 1612 |
+
},
|
| 1613 |
+
{
|
| 1614 |
+
"epoch": 24.33953033268102,
|
| 1615 |
+
"grad_norm": 0.014587494544684887,
|
| 1616 |
+
"learning_rate": 5.283757338551859e-07,
|
| 1617 |
+
"loss": 0.0067,
|
| 1618 |
+
"step": 99500
|
| 1619 |
+
},
|
| 1620 |
+
{
|
| 1621 |
+
"epoch": 24.461839530332682,
|
| 1622 |
+
"grad_norm": 0.7037560343742371,
|
| 1623 |
+
"learning_rate": 4.305283757338552e-07,
|
| 1624 |
+
"loss": 0.0059,
|
| 1625 |
+
"step": 100000
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 24.584148727984346,
|
| 1629 |
+
"grad_norm": 0.006079969462007284,
|
| 1630 |
+
"learning_rate": 3.326810176125245e-07,
|
| 1631 |
+
"loss": 0.0074,
|
| 1632 |
+
"step": 100500
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 24.70645792563601,
|
| 1636 |
+
"grad_norm": 1.3056970834732056,
|
| 1637 |
+
"learning_rate": 2.3483365949119375e-07,
|
| 1638 |
+
"loss": 0.0075,
|
| 1639 |
+
"step": 101000
|
| 1640 |
+
},
|
| 1641 |
+
{
|
| 1642 |
+
"epoch": 24.828767123287673,
|
| 1643 |
+
"grad_norm": 8.87458324432373,
|
| 1644 |
+
"learning_rate": 1.36986301369863e-07,
|
| 1645 |
+
"loss": 0.0065,
|
| 1646 |
+
"step": 101500
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 24.951076320939336,
|
| 1650 |
+
"grad_norm": 14.917941093444824,
|
| 1651 |
+
"learning_rate": 3.9138943248532294e-08,
|
| 1652 |
+
"loss": 0.0072,
|
| 1653 |
+
"step": 102000
|
| 1654 |
+
},
|
| 1655 |
+
{
|
| 1656 |
+
"epoch": 25.0,
|
| 1657 |
+
"eval_accuracy": 0.9311341199133156,
|
| 1658 |
+
"eval_loss": 0.49332907795906067,
|
| 1659 |
+
"eval_runtime": 79.9831,
|
| 1660 |
+
"eval_samples_per_second": 363.464,
|
| 1661 |
+
"eval_steps_per_second": 11.365,
|
| 1662 |
+
"step": 102200
|
| 1663 |
+
}
|
| 1664 |
+
],
|
| 1665 |
+
"logging_steps": 500,
|
| 1666 |
+
"max_steps": 102200,
|
| 1667 |
+
"num_input_tokens_seen": 0,
|
| 1668 |
+
"num_train_epochs": 25,
|
| 1669 |
+
"save_steps": 500,
|
| 1670 |
+
"stateful_callbacks": {
|
| 1671 |
+
"TrainerControl": {
|
| 1672 |
+
"args": {
|
| 1673 |
+
"should_epoch_stop": false,
|
| 1674 |
+
"should_evaluate": false,
|
| 1675 |
+
"should_log": false,
|
| 1676 |
+
"should_save": true,
|
| 1677 |
+
"should_training_stop": true
|
| 1678 |
+
},
|
| 1679 |
+
"attributes": {}
|
| 1680 |
+
}
|
| 1681 |
+
},
|
| 1682 |
+
"total_flos": 4.33360446554112e+17,
|
| 1683 |
+
"train_batch_size": 64,
|
| 1684 |
+
"trial_name": null,
|
| 1685 |
+
"trial_params": null
|
| 1686 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:25409054dd23ca427edffb1377223523e10e1ad97209f9a36a5eb35bb09ef8da
|
| 3 |
+
size 5112
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|