Add trained complexity classifier
Browse filesDistilBERT fine-tuned for question complexity classification
- checkpoint-1800/config.json +25 -0
- checkpoint-1800/model.safetensors +3 -0
- checkpoint-1800/optimizer.pt +3 -0
- checkpoint-1800/rng_state.pth +3 -0
- checkpoint-1800/scheduler.pt +3 -0
- checkpoint-1800/special_tokens_map.json +7 -0
- checkpoint-1800/tokenizer.json +0 -0
- checkpoint-1800/tokenizer_config.json +56 -0
- checkpoint-1800/trainer_state.json +1329 -0
- checkpoint-1800/training_args.bin +3 -0
- checkpoint-1800/vocab.txt +0 -0
- checkpoint-7200/config.json +25 -0
- checkpoint-7200/model.safetensors +3 -0
- checkpoint-7200/optimizer.pt +3 -0
- checkpoint-7200/rng_state.pth +3 -0
- checkpoint-7200/scheduler.pt +3 -0
- checkpoint-7200/special_tokens_map.json +7 -0
- checkpoint-7200/tokenizer.json +0 -0
- checkpoint-7200/tokenizer_config.json +56 -0
- checkpoint-7200/trainer_state.json +0 -0
- checkpoint-7200/training_args.bin +3 -0
- checkpoint-7200/vocab.txt +0 -0
- config.json +25 -0
- model.safetensors +3 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +56 -0
- vocab.txt +0 -0
checkpoint-1800/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"initializer_range": 0.02,
|
| 12 |
+
"max_position_embeddings": 512,
|
| 13 |
+
"model_type": "distilbert",
|
| 14 |
+
"n_heads": 12,
|
| 15 |
+
"n_layers": 6,
|
| 16 |
+
"pad_token_id": 0,
|
| 17 |
+
"problem_type": "single_label_classification",
|
| 18 |
+
"qa_dropout": 0.1,
|
| 19 |
+
"seq_classif_dropout": 0.2,
|
| 20 |
+
"sinusoidal_pos_embds": false,
|
| 21 |
+
"tie_weights_": true,
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.48.2",
|
| 24 |
+
"vocab_size": 30522
|
| 25 |
+
}
|
checkpoint-1800/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b6319bcf6a54195cbde5074dcc70cb6e404fc7cd1f87d52c076c374d2fe4f71a
|
| 3 |
+
size 267832560
|
checkpoint-1800/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c718f9b46407ee7f043d9e857fb8956cb77d781240e2ea75f819fcb167c5a96a
|
| 3 |
+
size 535727290
|
checkpoint-1800/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:296d6c2b5fde6d980700180dcc8992bcf9bbb0f68fcf1f07530f817621e9f52d
|
| 3 |
+
size 14244
|
checkpoint-1800/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d49396d59be4f8ba04c2d489be3b3355dbf96124f5f96e80722f70f60cd4caaf
|
| 3 |
+
size 1064
|
checkpoint-1800/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 |
+
}
|
checkpoint-1800/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1800/tokenizer_config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": false,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_lower_case": true,
|
| 47 |
+
"extra_special_tokens": {},
|
| 48 |
+
"mask_token": "[MASK]",
|
| 49 |
+
"model_max_length": 512,
|
| 50 |
+
"pad_token": "[PAD]",
|
| 51 |
+
"sep_token": "[SEP]",
|
| 52 |
+
"strip_accents": null,
|
| 53 |
+
"tokenize_chinese_chars": true,
|
| 54 |
+
"tokenizer_class": "DistilBertTokenizer",
|
| 55 |
+
"unk_token": "[UNK]"
|
| 56 |
+
}
|
checkpoint-1800/trainer_state.json
ADDED
|
@@ -0,0 +1,1329 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.663003663003663,
|
| 3 |
+
"best_model_checkpoint": "/home/grahama/workspace/experiments/complexity/model/checkpoint-1800",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1800,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.016666666666666666,
|
| 13 |
+
"grad_norm": 2.3482773303985596,
|
| 14 |
+
"learning_rate": 2.3809523809523811e-07,
|
| 15 |
+
"loss": 0.7019,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.03333333333333333,
|
| 20 |
+
"grad_norm": 1.3639864921569824,
|
| 21 |
+
"learning_rate": 4.7619047619047623e-07,
|
| 22 |
+
"loss": 0.7018,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.05,
|
| 27 |
+
"grad_norm": 2.9051921367645264,
|
| 28 |
+
"learning_rate": 7.142857142857143e-07,
|
| 29 |
+
"loss": 0.7151,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.06666666666666667,
|
| 34 |
+
"grad_norm": 1.0235440731048584,
|
| 35 |
+
"learning_rate": 9.523809523809525e-07,
|
| 36 |
+
"loss": 0.7034,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.08333333333333333,
|
| 41 |
+
"grad_norm": 2.771768808364868,
|
| 42 |
+
"learning_rate": 1.1904761904761906e-06,
|
| 43 |
+
"loss": 0.7006,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.1,
|
| 48 |
+
"grad_norm": 1.2305303812026978,
|
| 49 |
+
"learning_rate": 1.4285714285714286e-06,
|
| 50 |
+
"loss": 0.6995,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.11666666666666667,
|
| 55 |
+
"grad_norm": 2.2383503913879395,
|
| 56 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 57 |
+
"loss": 0.6994,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.13333333333333333,
|
| 62 |
+
"grad_norm": 1.0624160766601562,
|
| 63 |
+
"learning_rate": 1.904761904761905e-06,
|
| 64 |
+
"loss": 0.6998,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.15,
|
| 69 |
+
"grad_norm": 1.2018228769302368,
|
| 70 |
+
"learning_rate": 2.1428571428571427e-06,
|
| 71 |
+
"loss": 0.6974,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.16666666666666666,
|
| 76 |
+
"grad_norm": 1.0849791765213013,
|
| 77 |
+
"learning_rate": 2.380952380952381e-06,
|
| 78 |
+
"loss": 0.6875,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.18333333333333332,
|
| 83 |
+
"grad_norm": 1.291517972946167,
|
| 84 |
+
"learning_rate": 2.6190476190476192e-06,
|
| 85 |
+
"loss": 0.6865,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.2,
|
| 90 |
+
"grad_norm": 1.056863784790039,
|
| 91 |
+
"learning_rate": 2.8571428571428573e-06,
|
| 92 |
+
"loss": 0.6829,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.21666666666666667,
|
| 97 |
+
"grad_norm": 2.968041181564331,
|
| 98 |
+
"learning_rate": 3.0952380952380957e-06,
|
| 99 |
+
"loss": 0.6696,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.23333333333333334,
|
| 104 |
+
"grad_norm": 1.055704116821289,
|
| 105 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 106 |
+
"loss": 0.6783,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.25,
|
| 111 |
+
"grad_norm": 1.6586881875991821,
|
| 112 |
+
"learning_rate": 3.5714285714285718e-06,
|
| 113 |
+
"loss": 0.6702,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.26666666666666666,
|
| 118 |
+
"grad_norm": 2.3964574337005615,
|
| 119 |
+
"learning_rate": 3.80952380952381e-06,
|
| 120 |
+
"loss": 0.709,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.2833333333333333,
|
| 125 |
+
"grad_norm": 1.0744259357452393,
|
| 126 |
+
"learning_rate": 4.047619047619048e-06,
|
| 127 |
+
"loss": 0.7022,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.3,
|
| 132 |
+
"grad_norm": 1.071405053138733,
|
| 133 |
+
"learning_rate": 4.2857142857142855e-06,
|
| 134 |
+
"loss": 0.6747,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.31666666666666665,
|
| 139 |
+
"grad_norm": 2.1431033611297607,
|
| 140 |
+
"learning_rate": 4.523809523809524e-06,
|
| 141 |
+
"loss": 0.6623,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.3333333333333333,
|
| 146 |
+
"grad_norm": 1.301439881324768,
|
| 147 |
+
"learning_rate": 4.761904761904762e-06,
|
| 148 |
+
"loss": 0.667,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.35,
|
| 153 |
+
"grad_norm": 1.5243868827819824,
|
| 154 |
+
"learning_rate": 5e-06,
|
| 155 |
+
"loss": 0.6584,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.36666666666666664,
|
| 160 |
+
"grad_norm": 1.2769250869750977,
|
| 161 |
+
"learning_rate": 5.2380952380952384e-06,
|
| 162 |
+
"loss": 0.6898,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.38333333333333336,
|
| 167 |
+
"grad_norm": 1.274899959564209,
|
| 168 |
+
"learning_rate": 5.476190476190477e-06,
|
| 169 |
+
"loss": 0.6995,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.4,
|
| 174 |
+
"grad_norm": 1.658318281173706,
|
| 175 |
+
"learning_rate": 5.7142857142857145e-06,
|
| 176 |
+
"loss": 0.6705,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.4166666666666667,
|
| 181 |
+
"grad_norm": 2.4139277935028076,
|
| 182 |
+
"learning_rate": 5.9523809523809525e-06,
|
| 183 |
+
"loss": 0.6849,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.43333333333333335,
|
| 188 |
+
"grad_norm": 1.5813899040222168,
|
| 189 |
+
"learning_rate": 6.1904761904761914e-06,
|
| 190 |
+
"loss": 0.6663,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.45,
|
| 195 |
+
"grad_norm": 1.4048871994018555,
|
| 196 |
+
"learning_rate": 6.4285714285714295e-06,
|
| 197 |
+
"loss": 0.6542,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.4666666666666667,
|
| 202 |
+
"grad_norm": 1.7416518926620483,
|
| 203 |
+
"learning_rate": 6.666666666666667e-06,
|
| 204 |
+
"loss": 0.6793,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.48333333333333334,
|
| 209 |
+
"grad_norm": 1.7340086698532104,
|
| 210 |
+
"learning_rate": 6.9047619047619055e-06,
|
| 211 |
+
"loss": 0.6681,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.5,
|
| 216 |
+
"grad_norm": 1.4946725368499756,
|
| 217 |
+
"learning_rate": 7.1428571428571436e-06,
|
| 218 |
+
"loss": 0.6758,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.5166666666666667,
|
| 223 |
+
"grad_norm": 1.9904041290283203,
|
| 224 |
+
"learning_rate": 7.380952380952382e-06,
|
| 225 |
+
"loss": 0.6432,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.5333333333333333,
|
| 230 |
+
"grad_norm": 2.4469521045684814,
|
| 231 |
+
"learning_rate": 7.61904761904762e-06,
|
| 232 |
+
"loss": 0.6688,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.55,
|
| 237 |
+
"grad_norm": 2.104259967803955,
|
| 238 |
+
"learning_rate": 7.857142857142858e-06,
|
| 239 |
+
"loss": 0.6312,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.5666666666666667,
|
| 244 |
+
"grad_norm": 2.125582218170166,
|
| 245 |
+
"learning_rate": 8.095238095238097e-06,
|
| 246 |
+
"loss": 0.6712,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.5833333333333334,
|
| 251 |
+
"grad_norm": 1.9155024290084839,
|
| 252 |
+
"learning_rate": 8.333333333333334e-06,
|
| 253 |
+
"loss": 0.6766,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.6,
|
| 258 |
+
"grad_norm": 3.6578118801116943,
|
| 259 |
+
"learning_rate": 8.571428571428571e-06,
|
| 260 |
+
"loss": 0.6561,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.6166666666666667,
|
| 265 |
+
"grad_norm": 2.083104133605957,
|
| 266 |
+
"learning_rate": 8.80952380952381e-06,
|
| 267 |
+
"loss": 0.6374,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.6333333333333333,
|
| 272 |
+
"grad_norm": 4.414960861206055,
|
| 273 |
+
"learning_rate": 9.047619047619049e-06,
|
| 274 |
+
"loss": 0.6512,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.65,
|
| 279 |
+
"grad_norm": 2.075925588607788,
|
| 280 |
+
"learning_rate": 9.285714285714288e-06,
|
| 281 |
+
"loss": 0.6316,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.6666666666666666,
|
| 286 |
+
"grad_norm": 3.132584571838379,
|
| 287 |
+
"learning_rate": 9.523809523809525e-06,
|
| 288 |
+
"loss": 0.6424,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.6833333333333333,
|
| 293 |
+
"grad_norm": 2.2723052501678467,
|
| 294 |
+
"learning_rate": 9.761904761904762e-06,
|
| 295 |
+
"loss": 0.6395,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.7,
|
| 300 |
+
"grad_norm": 4.37326192855835,
|
| 301 |
+
"learning_rate": 1e-05,
|
| 302 |
+
"loss": 0.6519,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.7166666666666667,
|
| 307 |
+
"grad_norm": 3.2006280422210693,
|
| 308 |
+
"learning_rate": 1.0238095238095238e-05,
|
| 309 |
+
"loss": 0.6429,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.7333333333333333,
|
| 314 |
+
"grad_norm": 3.0944504737854004,
|
| 315 |
+
"learning_rate": 1.0476190476190477e-05,
|
| 316 |
+
"loss": 0.6756,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.75,
|
| 321 |
+
"grad_norm": 5.158036708831787,
|
| 322 |
+
"learning_rate": 1.0714285714285714e-05,
|
| 323 |
+
"loss": 0.6863,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.7666666666666667,
|
| 328 |
+
"grad_norm": 3.744805335998535,
|
| 329 |
+
"learning_rate": 1.0952380952380955e-05,
|
| 330 |
+
"loss": 0.6826,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.7833333333333333,
|
| 335 |
+
"grad_norm": 2.350935220718384,
|
| 336 |
+
"learning_rate": 1.1190476190476192e-05,
|
| 337 |
+
"loss": 0.6465,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.8,
|
| 342 |
+
"grad_norm": 2.391106367111206,
|
| 343 |
+
"learning_rate": 1.1428571428571429e-05,
|
| 344 |
+
"loss": 0.6494,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.8166666666666667,
|
| 349 |
+
"grad_norm": 2.6871049404144287,
|
| 350 |
+
"learning_rate": 1.1666666666666668e-05,
|
| 351 |
+
"loss": 0.614,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.8333333333333334,
|
| 356 |
+
"grad_norm": 2.496380567550659,
|
| 357 |
+
"learning_rate": 1.1904761904761905e-05,
|
| 358 |
+
"loss": 0.639,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.85,
|
| 363 |
+
"grad_norm": 3.1409785747528076,
|
| 364 |
+
"learning_rate": 1.2142857142857142e-05,
|
| 365 |
+
"loss": 0.6906,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.8666666666666667,
|
| 370 |
+
"grad_norm": 2.4331531524658203,
|
| 371 |
+
"learning_rate": 1.2380952380952383e-05,
|
| 372 |
+
"loss": 0.5914,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.8833333333333333,
|
| 377 |
+
"grad_norm": 2.956902265548706,
|
| 378 |
+
"learning_rate": 1.261904761904762e-05,
|
| 379 |
+
"loss": 0.7081,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.9,
|
| 384 |
+
"grad_norm": 2.1885573863983154,
|
| 385 |
+
"learning_rate": 1.2857142857142859e-05,
|
| 386 |
+
"loss": 0.6466,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.9166666666666666,
|
| 391 |
+
"grad_norm": 2.6431517601013184,
|
| 392 |
+
"learning_rate": 1.3095238095238096e-05,
|
| 393 |
+
"loss": 0.6631,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.9333333333333333,
|
| 398 |
+
"grad_norm": 2.1615426540374756,
|
| 399 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 400 |
+
"loss": 0.6735,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.95,
|
| 405 |
+
"grad_norm": 2.0207862854003906,
|
| 406 |
+
"learning_rate": 1.3571428571428574e-05,
|
| 407 |
+
"loss": 0.6325,
|
| 408 |
+
"step": 570
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.9666666666666667,
|
| 412 |
+
"grad_norm": 2.8899736404418945,
|
| 413 |
+
"learning_rate": 1.3809523809523811e-05,
|
| 414 |
+
"loss": 0.6519,
|
| 415 |
+
"step": 580
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.9833333333333333,
|
| 419 |
+
"grad_norm": 2.453443765640259,
|
| 420 |
+
"learning_rate": 1.4047619047619048e-05,
|
| 421 |
+
"loss": 0.6532,
|
| 422 |
+
"step": 590
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.0,
|
| 426 |
+
"grad_norm": 3.556234121322632,
|
| 427 |
+
"learning_rate": 1.4285714285714287e-05,
|
| 428 |
+
"loss": 0.5941,
|
| 429 |
+
"step": 600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.0,
|
| 433 |
+
"eval_accuracy": 0.6461538461538462,
|
| 434 |
+
"eval_loss": 0.6432204246520996,
|
| 435 |
+
"eval_runtime": 4.8353,
|
| 436 |
+
"eval_samples_per_second": 282.297,
|
| 437 |
+
"eval_steps_per_second": 17.786,
|
| 438 |
+
"step": 600
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 1.0166666666666666,
|
| 442 |
+
"grad_norm": 2.244147539138794,
|
| 443 |
+
"learning_rate": 1.4523809523809524e-05,
|
| 444 |
+
"loss": 0.6154,
|
| 445 |
+
"step": 610
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 1.0333333333333334,
|
| 449 |
+
"grad_norm": 2.3262217044830322,
|
| 450 |
+
"learning_rate": 1.4761904761904763e-05,
|
| 451 |
+
"loss": 0.5981,
|
| 452 |
+
"step": 620
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 1.05,
|
| 456 |
+
"grad_norm": 2.543795347213745,
|
| 457 |
+
"learning_rate": 1.5000000000000002e-05,
|
| 458 |
+
"loss": 0.6061,
|
| 459 |
+
"step": 630
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 1.0666666666666667,
|
| 463 |
+
"grad_norm": 3.5577542781829834,
|
| 464 |
+
"learning_rate": 1.523809523809524e-05,
|
| 465 |
+
"loss": 0.6432,
|
| 466 |
+
"step": 640
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 1.0833333333333333,
|
| 470 |
+
"grad_norm": 3.3030037879943848,
|
| 471 |
+
"learning_rate": 1.5476190476190476e-05,
|
| 472 |
+
"loss": 0.5663,
|
| 473 |
+
"step": 650
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 1.1,
|
| 477 |
+
"grad_norm": 2.2861194610595703,
|
| 478 |
+
"learning_rate": 1.5714285714285715e-05,
|
| 479 |
+
"loss": 0.6242,
|
| 480 |
+
"step": 660
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 1.1166666666666667,
|
| 484 |
+
"grad_norm": 3.9813036918640137,
|
| 485 |
+
"learning_rate": 1.5952380952380954e-05,
|
| 486 |
+
"loss": 0.6192,
|
| 487 |
+
"step": 670
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 1.1333333333333333,
|
| 491 |
+
"grad_norm": 5.103733062744141,
|
| 492 |
+
"learning_rate": 1.6190476190476193e-05,
|
| 493 |
+
"loss": 0.6288,
|
| 494 |
+
"step": 680
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 1.15,
|
| 498 |
+
"grad_norm": 4.313375949859619,
|
| 499 |
+
"learning_rate": 1.642857142857143e-05,
|
| 500 |
+
"loss": 0.5938,
|
| 501 |
+
"step": 690
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 1.1666666666666667,
|
| 505 |
+
"grad_norm": 3.633436918258667,
|
| 506 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 507 |
+
"loss": 0.6003,
|
| 508 |
+
"step": 700
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 1.1833333333333333,
|
| 512 |
+
"grad_norm": 3.060312509536743,
|
| 513 |
+
"learning_rate": 1.6904761904761906e-05,
|
| 514 |
+
"loss": 0.6214,
|
| 515 |
+
"step": 710
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 1.2,
|
| 519 |
+
"grad_norm": 3.6623165607452393,
|
| 520 |
+
"learning_rate": 1.7142857142857142e-05,
|
| 521 |
+
"loss": 0.6283,
|
| 522 |
+
"step": 720
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 1.2166666666666668,
|
| 526 |
+
"grad_norm": 2.556682586669922,
|
| 527 |
+
"learning_rate": 1.7380952380952384e-05,
|
| 528 |
+
"loss": 0.603,
|
| 529 |
+
"step": 730
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 1.2333333333333334,
|
| 533 |
+
"grad_norm": 3.2409424781799316,
|
| 534 |
+
"learning_rate": 1.761904761904762e-05,
|
| 535 |
+
"loss": 0.6463,
|
| 536 |
+
"step": 740
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 1.25,
|
| 540 |
+
"grad_norm": 3.6875534057617188,
|
| 541 |
+
"learning_rate": 1.785714285714286e-05,
|
| 542 |
+
"loss": 0.5864,
|
| 543 |
+
"step": 750
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 1.2666666666666666,
|
| 547 |
+
"grad_norm": 2.9428772926330566,
|
| 548 |
+
"learning_rate": 1.8095238095238097e-05,
|
| 549 |
+
"loss": 0.6349,
|
| 550 |
+
"step": 760
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 1.2833333333333332,
|
| 554 |
+
"grad_norm": 2.926853656768799,
|
| 555 |
+
"learning_rate": 1.8333333333333333e-05,
|
| 556 |
+
"loss": 0.6658,
|
| 557 |
+
"step": 770
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 1.3,
|
| 561 |
+
"grad_norm": 3.1792550086975098,
|
| 562 |
+
"learning_rate": 1.8571428571428575e-05,
|
| 563 |
+
"loss": 0.6263,
|
| 564 |
+
"step": 780
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 1.3166666666666667,
|
| 568 |
+
"grad_norm": 2.887868642807007,
|
| 569 |
+
"learning_rate": 1.880952380952381e-05,
|
| 570 |
+
"loss": 0.5737,
|
| 571 |
+
"step": 790
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 1.3333333333333333,
|
| 575 |
+
"grad_norm": 3.9215023517608643,
|
| 576 |
+
"learning_rate": 1.904761904761905e-05,
|
| 577 |
+
"loss": 0.5895,
|
| 578 |
+
"step": 800
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 1.35,
|
| 582 |
+
"grad_norm": 3.228184938430786,
|
| 583 |
+
"learning_rate": 1.928571428571429e-05,
|
| 584 |
+
"loss": 0.675,
|
| 585 |
+
"step": 810
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 1.3666666666666667,
|
| 589 |
+
"grad_norm": 2.5528361797332764,
|
| 590 |
+
"learning_rate": 1.9523809523809524e-05,
|
| 591 |
+
"loss": 0.612,
|
| 592 |
+
"step": 820
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 1.3833333333333333,
|
| 596 |
+
"grad_norm": 3.355010509490967,
|
| 597 |
+
"learning_rate": 1.9761904761904763e-05,
|
| 598 |
+
"loss": 0.6137,
|
| 599 |
+
"step": 830
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 1.4,
|
| 603 |
+
"grad_norm": 5.365524768829346,
|
| 604 |
+
"learning_rate": 2e-05,
|
| 605 |
+
"loss": 0.6278,
|
| 606 |
+
"step": 840
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 1.4166666666666667,
|
| 610 |
+
"grad_norm": 2.9840915203094482,
|
| 611 |
+
"learning_rate": 1.999991365731819e-05,
|
| 612 |
+
"loss": 0.6106,
|
| 613 |
+
"step": 850
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 1.4333333333333333,
|
| 617 |
+
"grad_norm": 3.0227549076080322,
|
| 618 |
+
"learning_rate": 1.999965463076377e-05,
|
| 619 |
+
"loss": 0.6136,
|
| 620 |
+
"step": 860
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 1.45,
|
| 624 |
+
"grad_norm": 4.935601711273193,
|
| 625 |
+
"learning_rate": 1.999922292480975e-05,
|
| 626 |
+
"loss": 0.6194,
|
| 627 |
+
"step": 870
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 1.4666666666666668,
|
| 631 |
+
"grad_norm": 2.947390556335449,
|
| 632 |
+
"learning_rate": 1.999861854691106e-05,
|
| 633 |
+
"loss": 0.6737,
|
| 634 |
+
"step": 880
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 1.4833333333333334,
|
| 638 |
+
"grad_norm": 3.2077040672302246,
|
| 639 |
+
"learning_rate": 1.999784150750442e-05,
|
| 640 |
+
"loss": 0.5906,
|
| 641 |
+
"step": 890
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 1.5,
|
| 645 |
+
"grad_norm": 2.1263020038604736,
|
| 646 |
+
"learning_rate": 1.9996891820008165e-05,
|
| 647 |
+
"loss": 0.6731,
|
| 648 |
+
"step": 900
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 1.5166666666666666,
|
| 652 |
+
"grad_norm": 3.1736927032470703,
|
| 653 |
+
"learning_rate": 1.9995769500822007e-05,
|
| 654 |
+
"loss": 0.5658,
|
| 655 |
+
"step": 910
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 1.5333333333333332,
|
| 659 |
+
"grad_norm": 3.766897201538086,
|
| 660 |
+
"learning_rate": 1.999447456932676e-05,
|
| 661 |
+
"loss": 0.6082,
|
| 662 |
+
"step": 920
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 1.55,
|
| 666 |
+
"grad_norm": 7.309018135070801,
|
| 667 |
+
"learning_rate": 1.9993007047883988e-05,
|
| 668 |
+
"loss": 0.5453,
|
| 669 |
+
"step": 930
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 1.5666666666666667,
|
| 673 |
+
"grad_norm": 6.418699264526367,
|
| 674 |
+
"learning_rate": 1.9991366961835643e-05,
|
| 675 |
+
"loss": 0.6428,
|
| 676 |
+
"step": 940
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 1.5833333333333335,
|
| 680 |
+
"grad_norm": 2.967588424682617,
|
| 681 |
+
"learning_rate": 1.9989554339503612e-05,
|
| 682 |
+
"loss": 0.6075,
|
| 683 |
+
"step": 950
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 1.6,
|
| 687 |
+
"grad_norm": 3.2338814735412598,
|
| 688 |
+
"learning_rate": 1.9987569212189224e-05,
|
| 689 |
+
"loss": 0.6513,
|
| 690 |
+
"step": 960
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 1.6166666666666667,
|
| 694 |
+
"grad_norm": 2.6137988567352295,
|
| 695 |
+
"learning_rate": 1.9985411614172728e-05,
|
| 696 |
+
"loss": 0.6457,
|
| 697 |
+
"step": 970
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 1.6333333333333333,
|
| 701 |
+
"grad_norm": 2.2644853591918945,
|
| 702 |
+
"learning_rate": 1.9983081582712684e-05,
|
| 703 |
+
"loss": 0.6146,
|
| 704 |
+
"step": 980
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 1.65,
|
| 708 |
+
"grad_norm": 3.580810785293579,
|
| 709 |
+
"learning_rate": 1.9980579158045322e-05,
|
| 710 |
+
"loss": 0.6434,
|
| 711 |
+
"step": 990
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 1.6666666666666665,
|
| 715 |
+
"grad_norm": 3.8415021896362305,
|
| 716 |
+
"learning_rate": 1.997790438338385e-05,
|
| 717 |
+
"loss": 0.6149,
|
| 718 |
+
"step": 1000
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 1.6833333333333333,
|
| 722 |
+
"grad_norm": 3.2889180183410645,
|
| 723 |
+
"learning_rate": 1.997505730491772e-05,
|
| 724 |
+
"loss": 0.6029,
|
| 725 |
+
"step": 1010
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 1.7,
|
| 729 |
+
"grad_norm": 3.0964486598968506,
|
| 730 |
+
"learning_rate": 1.9972037971811802e-05,
|
| 731 |
+
"loss": 0.6128,
|
| 732 |
+
"step": 1020
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 1.7166666666666668,
|
| 736 |
+
"grad_norm": 3.3787407875061035,
|
| 737 |
+
"learning_rate": 1.9968846436205566e-05,
|
| 738 |
+
"loss": 0.6478,
|
| 739 |
+
"step": 1030
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 1.7333333333333334,
|
| 743 |
+
"grad_norm": 3.098156452178955,
|
| 744 |
+
"learning_rate": 1.9965482753212154e-05,
|
| 745 |
+
"loss": 0.617,
|
| 746 |
+
"step": 1040
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 1.75,
|
| 750 |
+
"grad_norm": 3.4448206424713135,
|
| 751 |
+
"learning_rate": 1.9961946980917457e-05,
|
| 752 |
+
"loss": 0.6047,
|
| 753 |
+
"step": 1050
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 1.7666666666666666,
|
| 757 |
+
"grad_norm": 2.8418798446655273,
|
| 758 |
+
"learning_rate": 1.995823918037908e-05,
|
| 759 |
+
"loss": 0.6228,
|
| 760 |
+
"step": 1060
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 1.7833333333333332,
|
| 764 |
+
"grad_norm": 2.9955010414123535,
|
| 765 |
+
"learning_rate": 1.9954359415625313e-05,
|
| 766 |
+
"loss": 0.6033,
|
| 767 |
+
"step": 1070
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 1.8,
|
| 771 |
+
"grad_norm": 3.4788753986358643,
|
| 772 |
+
"learning_rate": 1.9950307753654016e-05,
|
| 773 |
+
"loss": 0.5801,
|
| 774 |
+
"step": 1080
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 1.8166666666666667,
|
| 778 |
+
"grad_norm": 2.713418483734131,
|
| 779 |
+
"learning_rate": 1.994608426443146e-05,
|
| 780 |
+
"loss": 0.5789,
|
| 781 |
+
"step": 1090
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 1.8333333333333335,
|
| 785 |
+
"grad_norm": 3.982531785964966,
|
| 786 |
+
"learning_rate": 1.994168902089112e-05,
|
| 787 |
+
"loss": 0.6497,
|
| 788 |
+
"step": 1100
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 1.85,
|
| 792 |
+
"grad_norm": 2.0478599071502686,
|
| 793 |
+
"learning_rate": 1.9937122098932428e-05,
|
| 794 |
+
"loss": 0.6031,
|
| 795 |
+
"step": 1110
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 1.8666666666666667,
|
| 799 |
+
"grad_norm": 5.567689418792725,
|
| 800 |
+
"learning_rate": 1.9932383577419432e-05,
|
| 801 |
+
"loss": 0.6487,
|
| 802 |
+
"step": 1120
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 1.8833333333333333,
|
| 806 |
+
"grad_norm": 2.5129470825195312,
|
| 807 |
+
"learning_rate": 1.9927473538179467e-05,
|
| 808 |
+
"loss": 0.5877,
|
| 809 |
+
"step": 1130
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 1.9,
|
| 813 |
+
"grad_norm": 2.4075396060943604,
|
| 814 |
+
"learning_rate": 1.9922392066001724e-05,
|
| 815 |
+
"loss": 0.6478,
|
| 816 |
+
"step": 1140
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 1.9166666666666665,
|
| 820 |
+
"grad_norm": 3.643559217453003,
|
| 821 |
+
"learning_rate": 1.9917139248635788e-05,
|
| 822 |
+
"loss": 0.625,
|
| 823 |
+
"step": 1150
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 1.9333333333333333,
|
| 827 |
+
"grad_norm": 3.3517959117889404,
|
| 828 |
+
"learning_rate": 1.991171517679013e-05,
|
| 829 |
+
"loss": 0.6502,
|
| 830 |
+
"step": 1160
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 1.95,
|
| 834 |
+
"grad_norm": 2.3340976238250732,
|
| 835 |
+
"learning_rate": 1.9906119944130527e-05,
|
| 836 |
+
"loss": 0.6354,
|
| 837 |
+
"step": 1170
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 1.9666666666666668,
|
| 841 |
+
"grad_norm": 2.662628173828125,
|
| 842 |
+
"learning_rate": 1.9900353647278466e-05,
|
| 843 |
+
"loss": 0.6282,
|
| 844 |
+
"step": 1180
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 1.9833333333333334,
|
| 848 |
+
"grad_norm": 3.3780112266540527,
|
| 849 |
+
"learning_rate": 1.9894416385809444e-05,
|
| 850 |
+
"loss": 0.6428,
|
| 851 |
+
"step": 1190
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 2.0,
|
| 855 |
+
"grad_norm": 1.9971626996994019,
|
| 856 |
+
"learning_rate": 1.9888308262251286e-05,
|
| 857 |
+
"loss": 0.6028,
|
| 858 |
+
"step": 1200
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 2.0,
|
| 862 |
+
"eval_accuracy": 0.6571428571428571,
|
| 863 |
+
"eval_loss": 0.6266757845878601,
|
| 864 |
+
"eval_runtime": 1.9428,
|
| 865 |
+
"eval_samples_per_second": 702.601,
|
| 866 |
+
"eval_steps_per_second": 44.266,
|
| 867 |
+
"step": 1200
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 2.0166666666666666,
|
| 871 |
+
"grad_norm": 2.3505845069885254,
|
| 872 |
+
"learning_rate": 1.9882029382082342e-05,
|
| 873 |
+
"loss": 0.5231,
|
| 874 |
+
"step": 1210
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 2.033333333333333,
|
| 878 |
+
"grad_norm": 3.543018102645874,
|
| 879 |
+
"learning_rate": 1.9875579853729677e-05,
|
| 880 |
+
"loss": 0.6002,
|
| 881 |
+
"step": 1220
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 2.05,
|
| 885 |
+
"grad_norm": 2.687847852706909,
|
| 886 |
+
"learning_rate": 1.9868959788567213e-05,
|
| 887 |
+
"loss": 0.5469,
|
| 888 |
+
"step": 1230
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 2.066666666666667,
|
| 892 |
+
"grad_norm": 4.663837909698486,
|
| 893 |
+
"learning_rate": 1.9862169300913784e-05,
|
| 894 |
+
"loss": 0.5582,
|
| 895 |
+
"step": 1240
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 2.0833333333333335,
|
| 899 |
+
"grad_norm": 4.063023090362549,
|
| 900 |
+
"learning_rate": 1.9855208508031173e-05,
|
| 901 |
+
"loss": 0.549,
|
| 902 |
+
"step": 1250
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 2.1,
|
| 906 |
+
"grad_norm": 5.092895984649658,
|
| 907 |
+
"learning_rate": 1.9848077530122083e-05,
|
| 908 |
+
"loss": 0.607,
|
| 909 |
+
"step": 1260
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 2.1166666666666667,
|
| 913 |
+
"grad_norm": 3.7984111309051514,
|
| 914 |
+
"learning_rate": 1.9840776490328067e-05,
|
| 915 |
+
"loss": 0.592,
|
| 916 |
+
"step": 1270
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 2.1333333333333333,
|
| 920 |
+
"grad_norm": 3.775224447250366,
|
| 921 |
+
"learning_rate": 1.9833305514727396e-05,
|
| 922 |
+
"loss": 0.5196,
|
| 923 |
+
"step": 1280
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 2.15,
|
| 927 |
+
"grad_norm": 5.623272895812988,
|
| 928 |
+
"learning_rate": 1.9825664732332886e-05,
|
| 929 |
+
"loss": 0.5283,
|
| 930 |
+
"step": 1290
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 2.1666666666666665,
|
| 934 |
+
"grad_norm": 4.856538772583008,
|
| 935 |
+
"learning_rate": 1.981785427508966e-05,
|
| 936 |
+
"loss": 0.518,
|
| 937 |
+
"step": 1300
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 2.183333333333333,
|
| 941 |
+
"grad_norm": 5.726941108703613,
|
| 942 |
+
"learning_rate": 1.9809874277872886e-05,
|
| 943 |
+
"loss": 0.5532,
|
| 944 |
+
"step": 1310
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 2.2,
|
| 948 |
+
"grad_norm": 5.120604515075684,
|
| 949 |
+
"learning_rate": 1.9801724878485438e-05,
|
| 950 |
+
"loss": 0.5172,
|
| 951 |
+
"step": 1320
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 2.216666666666667,
|
| 955 |
+
"grad_norm": 7.325108051300049,
|
| 956 |
+
"learning_rate": 1.9793406217655516e-05,
|
| 957 |
+
"loss": 0.5383,
|
| 958 |
+
"step": 1330
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 2.2333333333333334,
|
| 962 |
+
"grad_norm": 5.705660820007324,
|
| 963 |
+
"learning_rate": 1.9784918439034216e-05,
|
| 964 |
+
"loss": 0.4953,
|
| 965 |
+
"step": 1340
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 2.25,
|
| 969 |
+
"grad_norm": 6.060757637023926,
|
| 970 |
+
"learning_rate": 1.977626168919305e-05,
|
| 971 |
+
"loss": 0.5214,
|
| 972 |
+
"step": 1350
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 2.2666666666666666,
|
| 976 |
+
"grad_norm": 3.8743319511413574,
|
| 977 |
+
"learning_rate": 1.9767436117621416e-05,
|
| 978 |
+
"loss": 0.5362,
|
| 979 |
+
"step": 1360
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 2.283333333333333,
|
| 983 |
+
"grad_norm": 4.374029636383057,
|
| 984 |
+
"learning_rate": 1.975844187672402e-05,
|
| 985 |
+
"loss": 0.5604,
|
| 986 |
+
"step": 1370
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 2.3,
|
| 990 |
+
"grad_norm": 6.106194019317627,
|
| 991 |
+
"learning_rate": 1.9749279121818235e-05,
|
| 992 |
+
"loss": 0.5306,
|
| 993 |
+
"step": 1380
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 2.3166666666666664,
|
| 997 |
+
"grad_norm": 5.56414270401001,
|
| 998 |
+
"learning_rate": 1.9739948011131438e-05,
|
| 999 |
+
"loss": 0.5369,
|
| 1000 |
+
"step": 1390
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 2.3333333333333335,
|
| 1004 |
+
"grad_norm": 3.9365651607513428,
|
| 1005 |
+
"learning_rate": 1.973044870579824e-05,
|
| 1006 |
+
"loss": 0.526,
|
| 1007 |
+
"step": 1400
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 2.35,
|
| 1011 |
+
"grad_norm": 4.8061394691467285,
|
| 1012 |
+
"learning_rate": 1.9720781369857747e-05,
|
| 1013 |
+
"loss": 0.5264,
|
| 1014 |
+
"step": 1410
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 2.3666666666666667,
|
| 1018 |
+
"grad_norm": 5.648602485656738,
|
| 1019 |
+
"learning_rate": 1.9710946170250702e-05,
|
| 1020 |
+
"loss": 0.5488,
|
| 1021 |
+
"step": 1420
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 2.3833333333333333,
|
| 1025 |
+
"grad_norm": 4.482698917388916,
|
| 1026 |
+
"learning_rate": 1.9700943276816602e-05,
|
| 1027 |
+
"loss": 0.5014,
|
| 1028 |
+
"step": 1430
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 2.4,
|
| 1032 |
+
"grad_norm": 5.18354606628418,
|
| 1033 |
+
"learning_rate": 1.969077286229078e-05,
|
| 1034 |
+
"loss": 0.5383,
|
| 1035 |
+
"step": 1440
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 2.4166666666666665,
|
| 1039 |
+
"grad_norm": 5.076839923858643,
|
| 1040 |
+
"learning_rate": 1.9680435102301412e-05,
|
| 1041 |
+
"loss": 0.5387,
|
| 1042 |
+
"step": 1450
|
| 1043 |
+
},
|
| 1044 |
+
{
|
| 1045 |
+
"epoch": 2.4333333333333336,
|
| 1046 |
+
"grad_norm": 5.83772611618042,
|
| 1047 |
+
"learning_rate": 1.9669930175366474e-05,
|
| 1048 |
+
"loss": 0.4779,
|
| 1049 |
+
"step": 1460
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 2.45,
|
| 1053 |
+
"grad_norm": 5.16604471206665,
|
| 1054 |
+
"learning_rate": 1.9659258262890683e-05,
|
| 1055 |
+
"loss": 0.5521,
|
| 1056 |
+
"step": 1470
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 2.466666666666667,
|
| 1060 |
+
"grad_norm": 4.944518089294434,
|
| 1061 |
+
"learning_rate": 1.964841954916235e-05,
|
| 1062 |
+
"loss": 0.4844,
|
| 1063 |
+
"step": 1480
|
| 1064 |
+
},
|
| 1065 |
+
{
|
| 1066 |
+
"epoch": 2.4833333333333334,
|
| 1067 |
+
"grad_norm": 6.394619941711426,
|
| 1068 |
+
"learning_rate": 1.9637414221350198e-05,
|
| 1069 |
+
"loss": 0.567,
|
| 1070 |
+
"step": 1490
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 2.5,
|
| 1074 |
+
"grad_norm": 5.958597183227539,
|
| 1075 |
+
"learning_rate": 1.962624246950012e-05,
|
| 1076 |
+
"loss": 0.4951,
|
| 1077 |
+
"step": 1500
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 2.5166666666666666,
|
| 1081 |
+
"grad_norm": 4.636740207672119,
|
| 1082 |
+
"learning_rate": 1.9614904486531935e-05,
|
| 1083 |
+
"loss": 0.5123,
|
| 1084 |
+
"step": 1510
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 2.533333333333333,
|
| 1088 |
+
"grad_norm": 4.756173610687256,
|
| 1089 |
+
"learning_rate": 1.9603400468236e-05,
|
| 1090 |
+
"loss": 0.58,
|
| 1091 |
+
"step": 1520
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 2.55,
|
| 1095 |
+
"grad_norm": 3.969093084335327,
|
| 1096 |
+
"learning_rate": 1.9591730613269878e-05,
|
| 1097 |
+
"loss": 0.5517,
|
| 1098 |
+
"step": 1530
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 2.5666666666666664,
|
| 1102 |
+
"grad_norm": 4.5238237380981445,
|
| 1103 |
+
"learning_rate": 1.957989512315489e-05,
|
| 1104 |
+
"loss": 0.5167,
|
| 1105 |
+
"step": 1540
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 2.5833333333333335,
|
| 1109 |
+
"grad_norm": 5.810571670532227,
|
| 1110 |
+
"learning_rate": 1.9567894202272623e-05,
|
| 1111 |
+
"loss": 0.4826,
|
| 1112 |
+
"step": 1550
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 2.6,
|
| 1116 |
+
"grad_norm": 4.773239612579346,
|
| 1117 |
+
"learning_rate": 1.955572805786141e-05,
|
| 1118 |
+
"loss": 0.5443,
|
| 1119 |
+
"step": 1560
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 2.6166666666666667,
|
| 1123 |
+
"grad_norm": 4.351752281188965,
|
| 1124 |
+
"learning_rate": 1.9543396900012763e-05,
|
| 1125 |
+
"loss": 0.526,
|
| 1126 |
+
"step": 1570
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 2.6333333333333333,
|
| 1130 |
+
"grad_norm": 5.192873954772949,
|
| 1131 |
+
"learning_rate": 1.9530900941667733e-05,
|
| 1132 |
+
"loss": 0.5179,
|
| 1133 |
+
"step": 1580
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 2.65,
|
| 1137 |
+
"grad_norm": 5.123302936553955,
|
| 1138 |
+
"learning_rate": 1.9518240398613226e-05,
|
| 1139 |
+
"loss": 0.528,
|
| 1140 |
+
"step": 1590
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 2.6666666666666665,
|
| 1144 |
+
"grad_norm": 4.081309795379639,
|
| 1145 |
+
"learning_rate": 1.9505415489478293e-05,
|
| 1146 |
+
"loss": 0.592,
|
| 1147 |
+
"step": 1600
|
| 1148 |
+
},
|
| 1149 |
+
{
|
| 1150 |
+
"epoch": 2.6833333333333336,
|
| 1151 |
+
"grad_norm": 6.409515857696533,
|
| 1152 |
+
"learning_rate": 1.949242643573034e-05,
|
| 1153 |
+
"loss": 0.5388,
|
| 1154 |
+
"step": 1610
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 2.7,
|
| 1158 |
+
"grad_norm": 5.410388469696045,
|
| 1159 |
+
"learning_rate": 1.947927346167132e-05,
|
| 1160 |
+
"loss": 0.5233,
|
| 1161 |
+
"step": 1620
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 2.716666666666667,
|
| 1165 |
+
"grad_norm": 3.5235464572906494,
|
| 1166 |
+
"learning_rate": 1.9465956794433837e-05,
|
| 1167 |
+
"loss": 0.5285,
|
| 1168 |
+
"step": 1630
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 2.7333333333333334,
|
| 1172 |
+
"grad_norm": 5.431538105010986,
|
| 1173 |
+
"learning_rate": 1.945247666397725e-05,
|
| 1174 |
+
"loss": 0.4719,
|
| 1175 |
+
"step": 1640
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 2.75,
|
| 1179 |
+
"grad_norm": 6.913374900817871,
|
| 1180 |
+
"learning_rate": 1.9438833303083677e-05,
|
| 1181 |
+
"loss": 0.4603,
|
| 1182 |
+
"step": 1650
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 2.7666666666666666,
|
| 1186 |
+
"grad_norm": 6.34077262878418,
|
| 1187 |
+
"learning_rate": 1.9425026947353994e-05,
|
| 1188 |
+
"loss": 0.5288,
|
| 1189 |
+
"step": 1660
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 2.783333333333333,
|
| 1193 |
+
"grad_norm": 4.550387382507324,
|
| 1194 |
+
"learning_rate": 1.9411057835203756e-05,
|
| 1195 |
+
"loss": 0.4983,
|
| 1196 |
+
"step": 1670
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 2.8,
|
| 1200 |
+
"grad_norm": 4.740011692047119,
|
| 1201 |
+
"learning_rate": 1.9396926207859085e-05,
|
| 1202 |
+
"loss": 0.4939,
|
| 1203 |
+
"step": 1680
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 2.8166666666666664,
|
| 1207 |
+
"grad_norm": 5.84373664855957,
|
| 1208 |
+
"learning_rate": 1.9382632309352503e-05,
|
| 1209 |
+
"loss": 0.5498,
|
| 1210 |
+
"step": 1690
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 2.8333333333333335,
|
| 1214 |
+
"grad_norm": 4.3570027351379395,
|
| 1215 |
+
"learning_rate": 1.936817638651871e-05,
|
| 1216 |
+
"loss": 0.4716,
|
| 1217 |
+
"step": 1700
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 2.85,
|
| 1221 |
+
"grad_norm": 5.5916666984558105,
|
| 1222 |
+
"learning_rate": 1.935355868899034e-05,
|
| 1223 |
+
"loss": 0.5037,
|
| 1224 |
+
"step": 1710
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 2.8666666666666667,
|
| 1228 |
+
"grad_norm": 4.379909038543701,
|
| 1229 |
+
"learning_rate": 1.9338779469193638e-05,
|
| 1230 |
+
"loss": 0.5789,
|
| 1231 |
+
"step": 1720
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 2.8833333333333333,
|
| 1235 |
+
"grad_norm": 5.084365367889404,
|
| 1236 |
+
"learning_rate": 1.9323838982344092e-05,
|
| 1237 |
+
"loss": 0.5584,
|
| 1238 |
+
"step": 1730
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 2.9,
|
| 1242 |
+
"grad_norm": 6.741174221038818,
|
| 1243 |
+
"learning_rate": 1.9308737486442045e-05,
|
| 1244 |
+
"loss": 0.5089,
|
| 1245 |
+
"step": 1740
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 2.9166666666666665,
|
| 1249 |
+
"grad_norm": 6.0026469230651855,
|
| 1250 |
+
"learning_rate": 1.9293475242268224e-05,
|
| 1251 |
+
"loss": 0.4928,
|
| 1252 |
+
"step": 1750
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"epoch": 2.9333333333333336,
|
| 1256 |
+
"grad_norm": 6.163332939147949,
|
| 1257 |
+
"learning_rate": 1.9278052513379256e-05,
|
| 1258 |
+
"loss": 0.5397,
|
| 1259 |
+
"step": 1760
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 2.95,
|
| 1263 |
+
"grad_norm": 5.151211738586426,
|
| 1264 |
+
"learning_rate": 1.926246956610309e-05,
|
| 1265 |
+
"loss": 0.5085,
|
| 1266 |
+
"step": 1770
|
| 1267 |
+
},
|
| 1268 |
+
{
|
| 1269 |
+
"epoch": 2.966666666666667,
|
| 1270 |
+
"grad_norm": 5.902535915374756,
|
| 1271 |
+
"learning_rate": 1.9246726669534416e-05,
|
| 1272 |
+
"loss": 0.5619,
|
| 1273 |
+
"step": 1780
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 2.9833333333333334,
|
| 1277 |
+
"grad_norm": 4.26236629486084,
|
| 1278 |
+
"learning_rate": 1.923082409553002e-05,
|
| 1279 |
+
"loss": 0.4873,
|
| 1280 |
+
"step": 1790
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 3.0,
|
| 1284 |
+
"grad_norm": 9.270946502685547,
|
| 1285 |
+
"learning_rate": 1.921476211870408e-05,
|
| 1286 |
+
"loss": 0.5346,
|
| 1287 |
+
"step": 1800
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 3.0,
|
| 1291 |
+
"eval_accuracy": 0.663003663003663,
|
| 1292 |
+
"eval_loss": 0.6800894737243652,
|
| 1293 |
+
"eval_runtime": 2.0234,
|
| 1294 |
+
"eval_samples_per_second": 674.621,
|
| 1295 |
+
"eval_steps_per_second": 42.504,
|
| 1296 |
+
"step": 1800
|
| 1297 |
+
}
|
| 1298 |
+
],
|
| 1299 |
+
"logging_steps": 10,
|
| 1300 |
+
"max_steps": 8400,
|
| 1301 |
+
"num_input_tokens_seen": 0,
|
| 1302 |
+
"num_train_epochs": 14,
|
| 1303 |
+
"save_steps": 500,
|
| 1304 |
+
"stateful_callbacks": {
|
| 1305 |
+
"EarlyStoppingCallback": {
|
| 1306 |
+
"args": {
|
| 1307 |
+
"early_stopping_patience": 9,
|
| 1308 |
+
"early_stopping_threshold": 0.0
|
| 1309 |
+
},
|
| 1310 |
+
"attributes": {
|
| 1311 |
+
"early_stopping_patience_counter": 0
|
| 1312 |
+
}
|
| 1313 |
+
},
|
| 1314 |
+
"TrainerControl": {
|
| 1315 |
+
"args": {
|
| 1316 |
+
"should_epoch_stop": false,
|
| 1317 |
+
"should_evaluate": false,
|
| 1318 |
+
"should_log": false,
|
| 1319 |
+
"should_save": true,
|
| 1320 |
+
"should_training_stop": false
|
| 1321 |
+
},
|
| 1322 |
+
"attributes": {}
|
| 1323 |
+
}
|
| 1324 |
+
},
|
| 1325 |
+
"total_flos": 953665919774208.0,
|
| 1326 |
+
"train_batch_size": 16,
|
| 1327 |
+
"trial_name": null,
|
| 1328 |
+
"trial_params": null
|
| 1329 |
+
}
|
checkpoint-1800/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4733e371f005609eeefb32bbde1a2d14f17a4dfe340698358e00629b7cca8a02
|
| 3 |
+
size 5368
|
checkpoint-1800/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-7200/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"initializer_range": 0.02,
|
| 12 |
+
"max_position_embeddings": 512,
|
| 13 |
+
"model_type": "distilbert",
|
| 14 |
+
"n_heads": 12,
|
| 15 |
+
"n_layers": 6,
|
| 16 |
+
"pad_token_id": 0,
|
| 17 |
+
"problem_type": "single_label_classification",
|
| 18 |
+
"qa_dropout": 0.1,
|
| 19 |
+
"seq_classif_dropout": 0.2,
|
| 20 |
+
"sinusoidal_pos_embds": false,
|
| 21 |
+
"tie_weights_": true,
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.48.2",
|
| 24 |
+
"vocab_size": 30522
|
| 25 |
+
}
|
checkpoint-7200/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:83700abb52bd1d12fdf0ff9ee92e104797824a5a30533af589b1f2ce40732164
|
| 3 |
+
size 267832560
|
checkpoint-7200/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:01231ae74c278961c425db682eaa259f5ca8a5d6bdca633ea60feff08cf2aeb3
|
| 3 |
+
size 535727290
|
checkpoint-7200/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:062ce7f774653d1f3832a856d429b24c1205808af79b27ab89ce4e3e68927d42
|
| 3 |
+
size 14244
|
checkpoint-7200/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:026dfd4ed2e14f5eecfc818e03e207a5a9774cc2a49bd59c3911818bf95ed917
|
| 3 |
+
size 1064
|
checkpoint-7200/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 |
+
}
|
checkpoint-7200/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-7200/tokenizer_config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": false,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_lower_case": true,
|
| 47 |
+
"extra_special_tokens": {},
|
| 48 |
+
"mask_token": "[MASK]",
|
| 49 |
+
"model_max_length": 512,
|
| 50 |
+
"pad_token": "[PAD]",
|
| 51 |
+
"sep_token": "[SEP]",
|
| 52 |
+
"strip_accents": null,
|
| 53 |
+
"tokenize_chinese_chars": true,
|
| 54 |
+
"tokenizer_class": "DistilBertTokenizer",
|
| 55 |
+
"unk_token": "[UNK]"
|
| 56 |
+
}
|
checkpoint-7200/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-7200/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4733e371f005609eeefb32bbde1a2d14f17a4dfe340698358e00629b7cca8a02
|
| 3 |
+
size 5368
|
checkpoint-7200/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"initializer_range": 0.02,
|
| 12 |
+
"max_position_embeddings": 512,
|
| 13 |
+
"model_type": "distilbert",
|
| 14 |
+
"n_heads": 12,
|
| 15 |
+
"n_layers": 6,
|
| 16 |
+
"pad_token_id": 0,
|
| 17 |
+
"problem_type": "single_label_classification",
|
| 18 |
+
"qa_dropout": 0.1,
|
| 19 |
+
"seq_classif_dropout": 0.2,
|
| 20 |
+
"sinusoidal_pos_embds": false,
|
| 21 |
+
"tie_weights_": true,
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.48.2",
|
| 24 |
+
"vocab_size": 30522
|
| 25 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b6319bcf6a54195cbde5074dcc70cb6e404fc7cd1f87d52c076c374d2fe4f71a
|
| 3 |
+
size 267832560
|
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 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": false,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_lower_case": true,
|
| 47 |
+
"extra_special_tokens": {},
|
| 48 |
+
"mask_token": "[MASK]",
|
| 49 |
+
"model_max_length": 512,
|
| 50 |
+
"pad_token": "[PAD]",
|
| 51 |
+
"sep_token": "[SEP]",
|
| 52 |
+
"strip_accents": null,
|
| 53 |
+
"tokenize_chinese_chars": true,
|
| 54 |
+
"tokenizer_class": "DistilBertTokenizer",
|
| 55 |
+
"unk_token": "[UNK]"
|
| 56 |
+
}
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|