Upload folder using huggingface_hub
Browse files- checkpoint-3000/config.json +24 -0
- checkpoint-3000/model.safetensors +3 -0
- checkpoint-3000/optimizer.pt +3 -0
- checkpoint-3000/rng_state.pth +3 -0
- checkpoint-3000/scheduler.pt +3 -0
- checkpoint-3000/special_tokens_map.json +7 -0
- checkpoint-3000/tokenizer.json +0 -0
- checkpoint-3000/tokenizer_config.json +56 -0
- checkpoint-3000/trainer_state.json +2134 -0
- checkpoint-3000/training_args.bin +3 -0
- checkpoint-3000/vocab.txt +0 -0
- checkpoint-3482/config.json +24 -0
- checkpoint-3482/model.safetensors +3 -0
- checkpoint-3482/optimizer.pt +3 -0
- checkpoint-3482/rng_state.pth +3 -0
- checkpoint-3482/scheduler.pt +3 -0
- checkpoint-3482/special_tokens_map.json +7 -0
- checkpoint-3482/tokenizer.json +0 -0
- checkpoint-3482/tokenizer_config.json +56 -0
- checkpoint-3482/trainer_state.json +2470 -0
- checkpoint-3482/training_args.bin +3 -0
- checkpoint-3482/vocab.txt +0 -0
- config.json +24 -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-3000/config.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation": "gelu",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"DistilBertForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.1,
|
| 7 |
+
"dim": 768,
|
| 8 |
+
"dropout": 0.3,
|
| 9 |
+
"dtype": "float32",
|
| 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 |
+
"transformers_version": "4.57.1",
|
| 23 |
+
"vocab_size": 30522
|
| 24 |
+
}
|
checkpoint-3000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2a8eaf4876f4b852e5653d304488e78b7bae8529e8ea4dd2b83e0ca6b30e96e0
|
| 3 |
+
size 267832560
|
checkpoint-3000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:da2b0450e6a5519243581f5b565a53ee6b7973e0ac487ae19b0fd5a00438d3fe
|
| 3 |
+
size 535727290
|
checkpoint-3000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:053d6cdb71ddd1d4ac72326fefec638123d638f02de75d73320dddafa3a15da1
|
| 3 |
+
size 14244
|
checkpoint-3000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:16e807fd5448ed5db845d1bcf75950216827854b995e5193797c25bd4b557946
|
| 3 |
+
size 1064
|
checkpoint-3000/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-3000/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-3000/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-3000/trainer_state.json
ADDED
|
@@ -0,0 +1,2134 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.7231476163124642,
|
| 6 |
+
"eval_steps": 100,
|
| 7 |
+
"global_step": 3000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.005743825387708214,
|
| 14 |
+
"grad_norm": 1.664604902267456,
|
| 15 |
+
"learning_rate": 1.9948305571510626e-05,
|
| 16 |
+
"loss": 0.6766,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.011487650775416428,
|
| 21 |
+
"grad_norm": 1.8649318218231201,
|
| 22 |
+
"learning_rate": 1.9890867317633546e-05,
|
| 23 |
+
"loss": 0.6443,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.01723147616312464,
|
| 28 |
+
"grad_norm": 2.0676939487457275,
|
| 29 |
+
"learning_rate": 1.9833429063756463e-05,
|
| 30 |
+
"loss": 0.6516,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.022975301550832855,
|
| 35 |
+
"grad_norm": 2.077414035797119,
|
| 36 |
+
"learning_rate": 1.9775990809879383e-05,
|
| 37 |
+
"loss": 0.5468,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.02871912693854107,
|
| 42 |
+
"grad_norm": 2.4310338497161865,
|
| 43 |
+
"learning_rate": 1.97185525560023e-05,
|
| 44 |
+
"loss": 0.6289,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.03446295232624928,
|
| 49 |
+
"grad_norm": 2.605480670928955,
|
| 50 |
+
"learning_rate": 1.9661114302125216e-05,
|
| 51 |
+
"loss": 0.583,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.040206777713957496,
|
| 56 |
+
"grad_norm": 2.7097268104553223,
|
| 57 |
+
"learning_rate": 1.9603676048248136e-05,
|
| 58 |
+
"loss": 0.6029,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.04595060310166571,
|
| 63 |
+
"grad_norm": 4.125054359436035,
|
| 64 |
+
"learning_rate": 1.9546237794371053e-05,
|
| 65 |
+
"loss": 0.5545,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.051694428489373924,
|
| 70 |
+
"grad_norm": 4.592983245849609,
|
| 71 |
+
"learning_rate": 1.948879954049397e-05,
|
| 72 |
+
"loss": 0.5027,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.05743825387708214,
|
| 77 |
+
"grad_norm": 4.799990653991699,
|
| 78 |
+
"learning_rate": 1.9431361286616886e-05,
|
| 79 |
+
"loss": 0.4913,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.06318207926479034,
|
| 84 |
+
"grad_norm": 5.146159648895264,
|
| 85 |
+
"learning_rate": 1.9373923032739806e-05,
|
| 86 |
+
"loss": 0.5419,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.06892590465249857,
|
| 91 |
+
"grad_norm": 6.291711807250977,
|
| 92 |
+
"learning_rate": 1.9316484778862726e-05,
|
| 93 |
+
"loss": 0.478,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.07466973004020677,
|
| 98 |
+
"grad_norm": 4.076691627502441,
|
| 99 |
+
"learning_rate": 1.9259046524985643e-05,
|
| 100 |
+
"loss": 0.489,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.08041355542791499,
|
| 105 |
+
"grad_norm": 5.057150363922119,
|
| 106 |
+
"learning_rate": 1.920160827110856e-05,
|
| 107 |
+
"loss": 0.5435,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.0861573808156232,
|
| 112 |
+
"grad_norm": 3.5335211753845215,
|
| 113 |
+
"learning_rate": 1.914417001723148e-05,
|
| 114 |
+
"loss": 0.4445,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.09190120620333142,
|
| 119 |
+
"grad_norm": 6.209961414337158,
|
| 120 |
+
"learning_rate": 1.9086731763354396e-05,
|
| 121 |
+
"loss": 0.4516,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.09764503159103963,
|
| 126 |
+
"grad_norm": 7.1031341552734375,
|
| 127 |
+
"learning_rate": 1.9029293509477313e-05,
|
| 128 |
+
"loss": 0.576,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.10338885697874785,
|
| 133 |
+
"grad_norm": 4.959975719451904,
|
| 134 |
+
"learning_rate": 1.8971855255600233e-05,
|
| 135 |
+
"loss": 0.4841,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.10913268236645605,
|
| 140 |
+
"grad_norm": 4.904316425323486,
|
| 141 |
+
"learning_rate": 1.891441700172315e-05,
|
| 142 |
+
"loss": 0.5007,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.11487650775416428,
|
| 147 |
+
"grad_norm": 8.833961486816406,
|
| 148 |
+
"learning_rate": 1.8856978747846066e-05,
|
| 149 |
+
"loss": 0.5416,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.12062033314187248,
|
| 154 |
+
"grad_norm": 6.040881156921387,
|
| 155 |
+
"learning_rate": 1.8799540493968982e-05,
|
| 156 |
+
"loss": 0.5162,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.1263641585295807,
|
| 161 |
+
"grad_norm": 7.167252063751221,
|
| 162 |
+
"learning_rate": 1.8742102240091902e-05,
|
| 163 |
+
"loss": 0.4965,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.13210798391728892,
|
| 168 |
+
"grad_norm": 3.9503183364868164,
|
| 169 |
+
"learning_rate": 1.8684663986214822e-05,
|
| 170 |
+
"loss": 0.5129,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.13785180930499713,
|
| 175 |
+
"grad_norm": 4.403937339782715,
|
| 176 |
+
"learning_rate": 1.862722573233774e-05,
|
| 177 |
+
"loss": 0.5192,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.14359563469270534,
|
| 182 |
+
"grad_norm": 6.813930034637451,
|
| 183 |
+
"learning_rate": 1.8569787478460656e-05,
|
| 184 |
+
"loss": 0.4927,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.14933946008041354,
|
| 189 |
+
"grad_norm": 4.355352878570557,
|
| 190 |
+
"learning_rate": 1.8512349224583576e-05,
|
| 191 |
+
"loss": 0.4857,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.15508328546812178,
|
| 196 |
+
"grad_norm": 4.012150287628174,
|
| 197 |
+
"learning_rate": 1.8454910970706492e-05,
|
| 198 |
+
"loss": 0.3763,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.16082711085582999,
|
| 203 |
+
"grad_norm": 7.179994106292725,
|
| 204 |
+
"learning_rate": 1.839747271682941e-05,
|
| 205 |
+
"loss": 0.4625,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.1665709362435382,
|
| 210 |
+
"grad_norm": 3.701215982437134,
|
| 211 |
+
"learning_rate": 1.834003446295233e-05,
|
| 212 |
+
"loss": 0.4954,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.1723147616312464,
|
| 217 |
+
"grad_norm": 5.150369167327881,
|
| 218 |
+
"learning_rate": 1.8282596209075246e-05,
|
| 219 |
+
"loss": 0.4875,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.17805858701895463,
|
| 224 |
+
"grad_norm": 4.432060241699219,
|
| 225 |
+
"learning_rate": 1.8225157955198162e-05,
|
| 226 |
+
"loss": 0.5032,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.18380241240666284,
|
| 231 |
+
"grad_norm": 4.560088634490967,
|
| 232 |
+
"learning_rate": 1.816771970132108e-05,
|
| 233 |
+
"loss": 0.4443,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.18954623779437105,
|
| 238 |
+
"grad_norm": 6.000060558319092,
|
| 239 |
+
"learning_rate": 1.8110281447444e-05,
|
| 240 |
+
"loss": 0.4805,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.19529006318207925,
|
| 245 |
+
"grad_norm": 6.462589740753174,
|
| 246 |
+
"learning_rate": 1.805284319356692e-05,
|
| 247 |
+
"loss": 0.4009,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.2010338885697875,
|
| 252 |
+
"grad_norm": 4.6327104568481445,
|
| 253 |
+
"learning_rate": 1.7995404939689835e-05,
|
| 254 |
+
"loss": 0.5126,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.2067777139574957,
|
| 259 |
+
"grad_norm": 5.640307426452637,
|
| 260 |
+
"learning_rate": 1.7937966685812752e-05,
|
| 261 |
+
"loss": 0.4308,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.2125215393452039,
|
| 266 |
+
"grad_norm": 6.553674697875977,
|
| 267 |
+
"learning_rate": 1.7880528431935672e-05,
|
| 268 |
+
"loss": 0.4686,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.2182653647329121,
|
| 273 |
+
"grad_norm": 6.482476234436035,
|
| 274 |
+
"learning_rate": 1.782309017805859e-05,
|
| 275 |
+
"loss": 0.4214,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.22400919012062034,
|
| 280 |
+
"grad_norm": 6.897400379180908,
|
| 281 |
+
"learning_rate": 1.7765651924181505e-05,
|
| 282 |
+
"loss": 0.4827,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.22975301550832855,
|
| 287 |
+
"grad_norm": 3.935074806213379,
|
| 288 |
+
"learning_rate": 1.7708213670304425e-05,
|
| 289 |
+
"loss": 0.4212,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.23549684089603676,
|
| 294 |
+
"grad_norm": 5.766780853271484,
|
| 295 |
+
"learning_rate": 1.7650775416427342e-05,
|
| 296 |
+
"loss": 0.4421,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.24124066628374496,
|
| 301 |
+
"grad_norm": 8.834909439086914,
|
| 302 |
+
"learning_rate": 1.759333716255026e-05,
|
| 303 |
+
"loss": 0.4676,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.2469844916714532,
|
| 308 |
+
"grad_norm": 4.853402614593506,
|
| 309 |
+
"learning_rate": 1.7535898908673175e-05,
|
| 310 |
+
"loss": 0.4888,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.2527283170591614,
|
| 315 |
+
"grad_norm": 5.489238739013672,
|
| 316 |
+
"learning_rate": 1.7478460654796095e-05,
|
| 317 |
+
"loss": 0.3763,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.2584721424468696,
|
| 322 |
+
"grad_norm": 3.805001974105835,
|
| 323 |
+
"learning_rate": 1.7421022400919015e-05,
|
| 324 |
+
"loss": 0.5266,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.26421596783457785,
|
| 329 |
+
"grad_norm": 5.9350080490112305,
|
| 330 |
+
"learning_rate": 1.7363584147041932e-05,
|
| 331 |
+
"loss": 0.4053,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.269959793222286,
|
| 336 |
+
"grad_norm": 8.31826400756836,
|
| 337 |
+
"learning_rate": 1.730614589316485e-05,
|
| 338 |
+
"loss": 0.4775,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.27570361860999426,
|
| 343 |
+
"grad_norm": 5.709866046905518,
|
| 344 |
+
"learning_rate": 1.724870763928777e-05,
|
| 345 |
+
"loss": 0.3681,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.2814474439977025,
|
| 350 |
+
"grad_norm": 2.6657683849334717,
|
| 351 |
+
"learning_rate": 1.7191269385410685e-05,
|
| 352 |
+
"loss": 0.3913,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.2871912693854107,
|
| 357 |
+
"grad_norm": 8.942373275756836,
|
| 358 |
+
"learning_rate": 1.71338311315336e-05,
|
| 359 |
+
"loss": 0.4454,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.2929350947731189,
|
| 364 |
+
"grad_norm": 7.566854476928711,
|
| 365 |
+
"learning_rate": 1.707639287765652e-05,
|
| 366 |
+
"loss": 0.4704,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.2986789201608271,
|
| 371 |
+
"grad_norm": 5.819560527801514,
|
| 372 |
+
"learning_rate": 1.7018954623779438e-05,
|
| 373 |
+
"loss": 0.4245,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.3044227455485353,
|
| 378 |
+
"grad_norm": 7.269554138183594,
|
| 379 |
+
"learning_rate": 1.6961516369902355e-05,
|
| 380 |
+
"loss": 0.523,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.31016657093624356,
|
| 385 |
+
"grad_norm": 4.569669246673584,
|
| 386 |
+
"learning_rate": 1.690407811602527e-05,
|
| 387 |
+
"loss": 0.5053,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.31591039632395174,
|
| 392 |
+
"grad_norm": 6.8373332023620605,
|
| 393 |
+
"learning_rate": 1.684663986214819e-05,
|
| 394 |
+
"loss": 0.4592,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.32165422171165997,
|
| 399 |
+
"grad_norm": 7.736797332763672,
|
| 400 |
+
"learning_rate": 1.678920160827111e-05,
|
| 401 |
+
"loss": 0.4756,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.3273980470993682,
|
| 406 |
+
"grad_norm": 5.857370853424072,
|
| 407 |
+
"learning_rate": 1.6731763354394028e-05,
|
| 408 |
+
"loss": 0.4906,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.3331418724870764,
|
| 413 |
+
"grad_norm": 4.2734785079956055,
|
| 414 |
+
"learning_rate": 1.6674325100516945e-05,
|
| 415 |
+
"loss": 0.4623,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.3388856978747846,
|
| 420 |
+
"grad_norm": 7.859753131866455,
|
| 421 |
+
"learning_rate": 1.6616886846639865e-05,
|
| 422 |
+
"loss": 0.4568,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.3446295232624928,
|
| 427 |
+
"grad_norm": 4.936352252960205,
|
| 428 |
+
"learning_rate": 1.655944859276278e-05,
|
| 429 |
+
"loss": 0.3774,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.35037334865020103,
|
| 434 |
+
"grad_norm": 7.164617538452148,
|
| 435 |
+
"learning_rate": 1.6502010338885698e-05,
|
| 436 |
+
"loss": 0.5373,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.35611717403790927,
|
| 441 |
+
"grad_norm": 7.279328346252441,
|
| 442 |
+
"learning_rate": 1.6444572085008618e-05,
|
| 443 |
+
"loss": 0.4183,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.36186099942561745,
|
| 448 |
+
"grad_norm": 8.805562973022461,
|
| 449 |
+
"learning_rate": 1.6387133831131535e-05,
|
| 450 |
+
"loss": 0.4232,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.3676048248133257,
|
| 455 |
+
"grad_norm": 6.055676460266113,
|
| 456 |
+
"learning_rate": 1.632969557725445e-05,
|
| 457 |
+
"loss": 0.4316,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.3733486502010339,
|
| 462 |
+
"grad_norm": 5.835373401641846,
|
| 463 |
+
"learning_rate": 1.627225732337737e-05,
|
| 464 |
+
"loss": 0.4573,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.3790924755887421,
|
| 469 |
+
"grad_norm": 2.7914931774139404,
|
| 470 |
+
"learning_rate": 1.6214819069500288e-05,
|
| 471 |
+
"loss": 0.3691,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.38483630097645033,
|
| 476 |
+
"grad_norm": 5.780324459075928,
|
| 477 |
+
"learning_rate": 1.6157380815623208e-05,
|
| 478 |
+
"loss": 0.3761,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.3905801263641585,
|
| 483 |
+
"grad_norm": 5.7374348640441895,
|
| 484 |
+
"learning_rate": 1.6099942561746125e-05,
|
| 485 |
+
"loss": 0.4276,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.39632395175186674,
|
| 490 |
+
"grad_norm": 4.771572589874268,
|
| 491 |
+
"learning_rate": 1.604250430786904e-05,
|
| 492 |
+
"loss": 0.4092,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.402067777139575,
|
| 497 |
+
"grad_norm": 7.6168928146362305,
|
| 498 |
+
"learning_rate": 1.598506605399196e-05,
|
| 499 |
+
"loss": 0.4651,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.40781160252728316,
|
| 504 |
+
"grad_norm": 6.272294998168945,
|
| 505 |
+
"learning_rate": 1.5927627800114878e-05,
|
| 506 |
+
"loss": 0.4107,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.4135554279149914,
|
| 511 |
+
"grad_norm": 8.060046195983887,
|
| 512 |
+
"learning_rate": 1.5870189546237794e-05,
|
| 513 |
+
"loss": 0.4117,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.41929925330269957,
|
| 518 |
+
"grad_norm": 7.290645122528076,
|
| 519 |
+
"learning_rate": 1.5812751292360714e-05,
|
| 520 |
+
"loss": 0.5005,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.4250430786904078,
|
| 525 |
+
"grad_norm": 8.02639102935791,
|
| 526 |
+
"learning_rate": 1.575531303848363e-05,
|
| 527 |
+
"loss": 0.4561,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.43078690407811604,
|
| 532 |
+
"grad_norm": 8.455704689025879,
|
| 533 |
+
"learning_rate": 1.5697874784606548e-05,
|
| 534 |
+
"loss": 0.4592,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.4365307294658242,
|
| 539 |
+
"grad_norm": 2.9264323711395264,
|
| 540 |
+
"learning_rate": 1.5640436530729468e-05,
|
| 541 |
+
"loss": 0.4532,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.44227455485353245,
|
| 546 |
+
"grad_norm": 5.246438980102539,
|
| 547 |
+
"learning_rate": 1.5582998276852384e-05,
|
| 548 |
+
"loss": 0.4931,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.4480183802412407,
|
| 553 |
+
"grad_norm": 3.869628667831421,
|
| 554 |
+
"learning_rate": 1.5525560022975304e-05,
|
| 555 |
+
"loss": 0.3476,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.45376220562894887,
|
| 560 |
+
"grad_norm": 4.742978572845459,
|
| 561 |
+
"learning_rate": 1.546812176909822e-05,
|
| 562 |
+
"loss": 0.4274,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.4595060310166571,
|
| 567 |
+
"grad_norm": 6.9326300621032715,
|
| 568 |
+
"learning_rate": 1.5410683515221138e-05,
|
| 569 |
+
"loss": 0.5299,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.4652498564043653,
|
| 574 |
+
"grad_norm": 6.16196346282959,
|
| 575 |
+
"learning_rate": 1.5353245261344058e-05,
|
| 576 |
+
"loss": 0.4805,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.4709936817920735,
|
| 581 |
+
"grad_norm": 3.637753486633301,
|
| 582 |
+
"learning_rate": 1.5295807007466974e-05,
|
| 583 |
+
"loss": 0.4393,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.47673750717978175,
|
| 588 |
+
"grad_norm": 3.867966413497925,
|
| 589 |
+
"learning_rate": 1.523836875358989e-05,
|
| 590 |
+
"loss": 0.4358,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.48248133256748993,
|
| 595 |
+
"grad_norm": 4.753304481506348,
|
| 596 |
+
"learning_rate": 1.5180930499712809e-05,
|
| 597 |
+
"loss": 0.4737,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.48822515795519816,
|
| 602 |
+
"grad_norm": 7.7434515953063965,
|
| 603 |
+
"learning_rate": 1.5123492245835727e-05,
|
| 604 |
+
"loss": 0.4466,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.4939689833429064,
|
| 609 |
+
"grad_norm": 6.001241683959961,
|
| 610 |
+
"learning_rate": 1.5066053991958644e-05,
|
| 611 |
+
"loss": 0.4016,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.4997128087306146,
|
| 616 |
+
"grad_norm": 7.710599422454834,
|
| 617 |
+
"learning_rate": 1.5008615738081564e-05,
|
| 618 |
+
"loss": 0.4758,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.5054566341183228,
|
| 623 |
+
"grad_norm": 5.139091491699219,
|
| 624 |
+
"learning_rate": 1.4951177484204482e-05,
|
| 625 |
+
"loss": 0.4019,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.511200459506031,
|
| 630 |
+
"grad_norm": 9.428766250610352,
|
| 631 |
+
"learning_rate": 1.4893739230327399e-05,
|
| 632 |
+
"loss": 0.3742,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.5169442848937392,
|
| 637 |
+
"grad_norm": 9.071430206298828,
|
| 638 |
+
"learning_rate": 1.4836300976450317e-05,
|
| 639 |
+
"loss": 0.4183,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.5226881102814475,
|
| 644 |
+
"grad_norm": 10.34457015991211,
|
| 645 |
+
"learning_rate": 1.4778862722573236e-05,
|
| 646 |
+
"loss": 0.437,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.5284319356691557,
|
| 651 |
+
"grad_norm": 7.01099967956543,
|
| 652 |
+
"learning_rate": 1.4721424468696152e-05,
|
| 653 |
+
"loss": 0.4421,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.5341757610568638,
|
| 658 |
+
"grad_norm": 3.454484701156616,
|
| 659 |
+
"learning_rate": 1.466398621481907e-05,
|
| 660 |
+
"loss": 0.4283,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.539919586444572,
|
| 665 |
+
"grad_norm": 5.86787223815918,
|
| 666 |
+
"learning_rate": 1.4606547960941987e-05,
|
| 667 |
+
"loss": 0.4094,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.5456634118322803,
|
| 672 |
+
"grad_norm": 8.41640567779541,
|
| 673 |
+
"learning_rate": 1.4549109707064906e-05,
|
| 674 |
+
"loss": 0.414,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.5514072372199885,
|
| 679 |
+
"grad_norm": 6.936724662780762,
|
| 680 |
+
"learning_rate": 1.4491671453187824e-05,
|
| 681 |
+
"loss": 0.3936,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.5571510626076968,
|
| 686 |
+
"grad_norm": 7.984805583953857,
|
| 687 |
+
"learning_rate": 1.4434233199310744e-05,
|
| 688 |
+
"loss": 0.4071,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.562894887995405,
|
| 693 |
+
"grad_norm": 4.381742477416992,
|
| 694 |
+
"learning_rate": 1.437679494543366e-05,
|
| 695 |
+
"loss": 0.3393,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.5686387133831131,
|
| 700 |
+
"grad_norm": 7.115252494812012,
|
| 701 |
+
"learning_rate": 1.4319356691556579e-05,
|
| 702 |
+
"loss": 0.5616,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.5743825387708213,
|
| 707 |
+
"grad_norm": 6.60570764541626,
|
| 708 |
+
"learning_rate": 1.4261918437679495e-05,
|
| 709 |
+
"loss": 0.4633,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.5801263641585296,
|
| 714 |
+
"grad_norm": 9.552061080932617,
|
| 715 |
+
"learning_rate": 1.4204480183802414e-05,
|
| 716 |
+
"loss": 0.4973,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.5858701895462378,
|
| 721 |
+
"grad_norm": 3.6467580795288086,
|
| 722 |
+
"learning_rate": 1.4147041929925332e-05,
|
| 723 |
+
"loss": 0.3958,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.591614014933946,
|
| 728 |
+
"grad_norm": 2.663799524307251,
|
| 729 |
+
"learning_rate": 1.4089603676048249e-05,
|
| 730 |
+
"loss": 0.3402,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.5973578403216542,
|
| 735 |
+
"grad_norm": 7.468900203704834,
|
| 736 |
+
"learning_rate": 1.4032165422171167e-05,
|
| 737 |
+
"loss": 0.3341,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.6031016657093624,
|
| 742 |
+
"grad_norm": 10.396268844604492,
|
| 743 |
+
"learning_rate": 1.3974727168294084e-05,
|
| 744 |
+
"loss": 0.5122,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.6088454910970706,
|
| 749 |
+
"grad_norm": 7.79450798034668,
|
| 750 |
+
"learning_rate": 1.3917288914417002e-05,
|
| 751 |
+
"loss": 0.5581,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.6145893164847789,
|
| 756 |
+
"grad_norm": 8.077397346496582,
|
| 757 |
+
"learning_rate": 1.385985066053992e-05,
|
| 758 |
+
"loss": 0.4247,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.6203331418724871,
|
| 763 |
+
"grad_norm": 8.327542304992676,
|
| 764 |
+
"learning_rate": 1.380241240666284e-05,
|
| 765 |
+
"loss": 0.4122,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.6260769672601952,
|
| 770 |
+
"grad_norm": 7.940774917602539,
|
| 771 |
+
"learning_rate": 1.3744974152785757e-05,
|
| 772 |
+
"loss": 0.5199,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.6318207926479035,
|
| 777 |
+
"grad_norm": 5.148271560668945,
|
| 778 |
+
"learning_rate": 1.3687535898908675e-05,
|
| 779 |
+
"loss": 0.4534,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.6375646180356117,
|
| 784 |
+
"grad_norm": 7.042996883392334,
|
| 785 |
+
"learning_rate": 1.3630097645031592e-05,
|
| 786 |
+
"loss": 0.4737,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.6433084434233199,
|
| 791 |
+
"grad_norm": 5.131284236907959,
|
| 792 |
+
"learning_rate": 1.357265939115451e-05,
|
| 793 |
+
"loss": 0.3637,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.6490522688110282,
|
| 798 |
+
"grad_norm": 10.73865795135498,
|
| 799 |
+
"learning_rate": 1.3515221137277428e-05,
|
| 800 |
+
"loss": 0.4152,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.6547960941987364,
|
| 805 |
+
"grad_norm": 6.061016082763672,
|
| 806 |
+
"learning_rate": 1.3457782883400345e-05,
|
| 807 |
+
"loss": 0.343,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.6605399195864445,
|
| 812 |
+
"grad_norm": 11.200201988220215,
|
| 813 |
+
"learning_rate": 1.3400344629523263e-05,
|
| 814 |
+
"loss": 0.4781,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.6662837449741528,
|
| 819 |
+
"grad_norm": 6.987539768218994,
|
| 820 |
+
"learning_rate": 1.334290637564618e-05,
|
| 821 |
+
"loss": 0.4046,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.672027570361861,
|
| 826 |
+
"grad_norm": 7.3787713050842285,
|
| 827 |
+
"learning_rate": 1.3285468121769098e-05,
|
| 828 |
+
"loss": 0.4136,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.6777713957495692,
|
| 833 |
+
"grad_norm": 8.829427719116211,
|
| 834 |
+
"learning_rate": 1.3228029867892018e-05,
|
| 835 |
+
"loss": 0.3807,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.6835152211372775,
|
| 840 |
+
"grad_norm": 9.648842811584473,
|
| 841 |
+
"learning_rate": 1.3170591614014937e-05,
|
| 842 |
+
"loss": 0.3273,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.6892590465249856,
|
| 847 |
+
"grad_norm": 7.307587146759033,
|
| 848 |
+
"learning_rate": 1.3113153360137853e-05,
|
| 849 |
+
"loss": 0.3351,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.6950028719126938,
|
| 854 |
+
"grad_norm": 6.445584297180176,
|
| 855 |
+
"learning_rate": 1.3055715106260772e-05,
|
| 856 |
+
"loss": 0.4776,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.7007466973004021,
|
| 861 |
+
"grad_norm": 7.078349590301514,
|
| 862 |
+
"learning_rate": 1.2998276852383688e-05,
|
| 863 |
+
"loss": 0.4438,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.7064905226881103,
|
| 868 |
+
"grad_norm": 9.63571834564209,
|
| 869 |
+
"learning_rate": 1.2940838598506606e-05,
|
| 870 |
+
"loss": 0.4027,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.7122343480758185,
|
| 875 |
+
"grad_norm": 3.2861080169677734,
|
| 876 |
+
"learning_rate": 1.2883400344629525e-05,
|
| 877 |
+
"loss": 0.3334,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.7179781734635267,
|
| 882 |
+
"grad_norm": 7.433917999267578,
|
| 883 |
+
"learning_rate": 1.2825962090752441e-05,
|
| 884 |
+
"loss": 0.4418,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.7237219988512349,
|
| 889 |
+
"grad_norm": 7.511765003204346,
|
| 890 |
+
"learning_rate": 1.276852383687536e-05,
|
| 891 |
+
"loss": 0.3693,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.7294658242389431,
|
| 896 |
+
"grad_norm": 9.38160228729248,
|
| 897 |
+
"learning_rate": 1.2711085582998276e-05,
|
| 898 |
+
"loss": 0.3989,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.7352096496266514,
|
| 903 |
+
"grad_norm": 2.012756586074829,
|
| 904 |
+
"learning_rate": 1.2653647329121195e-05,
|
| 905 |
+
"loss": 0.3867,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.7409534750143596,
|
| 910 |
+
"grad_norm": 6.777096748352051,
|
| 911 |
+
"learning_rate": 1.2596209075244115e-05,
|
| 912 |
+
"loss": 0.5528,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.7466973004020678,
|
| 917 |
+
"grad_norm": 6.928145885467529,
|
| 918 |
+
"learning_rate": 1.2538770821367033e-05,
|
| 919 |
+
"loss": 0.3403,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.752441125789776,
|
| 924 |
+
"grad_norm": 7.99967622756958,
|
| 925 |
+
"learning_rate": 1.248133256748995e-05,
|
| 926 |
+
"loss": 0.5006,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.7581849511774842,
|
| 931 |
+
"grad_norm": 4.8364033699035645,
|
| 932 |
+
"learning_rate": 1.2423894313612868e-05,
|
| 933 |
+
"loss": 0.4266,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.7639287765651924,
|
| 938 |
+
"grad_norm": 4.017831802368164,
|
| 939 |
+
"learning_rate": 1.2366456059735785e-05,
|
| 940 |
+
"loss": 0.3444,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.7696726019529007,
|
| 945 |
+
"grad_norm": 5.27449893951416,
|
| 946 |
+
"learning_rate": 1.2309017805858703e-05,
|
| 947 |
+
"loss": 0.3962,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.7754164273406089,
|
| 952 |
+
"grad_norm": 7.989853858947754,
|
| 953 |
+
"learning_rate": 1.2251579551981621e-05,
|
| 954 |
+
"loss": 0.4172,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.781160252728317,
|
| 959 |
+
"grad_norm": 5.878440856933594,
|
| 960 |
+
"learning_rate": 1.2194141298104538e-05,
|
| 961 |
+
"loss": 0.4723,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.7869040781160253,
|
| 966 |
+
"grad_norm": 9.140411376953125,
|
| 967 |
+
"learning_rate": 1.2136703044227456e-05,
|
| 968 |
+
"loss": 0.4012,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.7926479035037335,
|
| 973 |
+
"grad_norm": 3.9119012355804443,
|
| 974 |
+
"learning_rate": 1.2079264790350373e-05,
|
| 975 |
+
"loss": 0.3545,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.7983917288914417,
|
| 980 |
+
"grad_norm": 6.248933792114258,
|
| 981 |
+
"learning_rate": 1.2021826536473291e-05,
|
| 982 |
+
"loss": 0.4787,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.80413555427915,
|
| 987 |
+
"grad_norm": 8.478959083557129,
|
| 988 |
+
"learning_rate": 1.1964388282596211e-05,
|
| 989 |
+
"loss": 0.4077,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.8098793796668581,
|
| 994 |
+
"grad_norm": 6.234384059906006,
|
| 995 |
+
"learning_rate": 1.190695002871913e-05,
|
| 996 |
+
"loss": 0.4044,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.8156232050545663,
|
| 1001 |
+
"grad_norm": 5.093031883239746,
|
| 1002 |
+
"learning_rate": 1.1849511774842046e-05,
|
| 1003 |
+
"loss": 0.3298,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.8213670304422745,
|
| 1008 |
+
"grad_norm": 5.755350112915039,
|
| 1009 |
+
"learning_rate": 1.1792073520964964e-05,
|
| 1010 |
+
"loss": 0.4099,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.8271108558299828,
|
| 1015 |
+
"grad_norm": 9.269704818725586,
|
| 1016 |
+
"learning_rate": 1.1734635267087881e-05,
|
| 1017 |
+
"loss": 0.366,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.832854681217691,
|
| 1022 |
+
"grad_norm": 4.977533340454102,
|
| 1023 |
+
"learning_rate": 1.16771970132108e-05,
|
| 1024 |
+
"loss": 0.3156,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.8385985066053991,
|
| 1029 |
+
"grad_norm": 6.767063140869141,
|
| 1030 |
+
"learning_rate": 1.1619758759333718e-05,
|
| 1031 |
+
"loss": 0.3966,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.8443423319931074,
|
| 1036 |
+
"grad_norm": 6.855627536773682,
|
| 1037 |
+
"learning_rate": 1.1562320505456634e-05,
|
| 1038 |
+
"loss": 0.3488,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.8500861573808156,
|
| 1043 |
+
"grad_norm": 3.408679723739624,
|
| 1044 |
+
"learning_rate": 1.1504882251579552e-05,
|
| 1045 |
+
"loss": 0.4016,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.8558299827685238,
|
| 1050 |
+
"grad_norm": 8.43376636505127,
|
| 1051 |
+
"learning_rate": 1.1447443997702469e-05,
|
| 1052 |
+
"loss": 0.3951,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.8615738081562321,
|
| 1057 |
+
"grad_norm": 7.106573104858398,
|
| 1058 |
+
"learning_rate": 1.1390005743825389e-05,
|
| 1059 |
+
"loss": 0.3056,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.8673176335439403,
|
| 1064 |
+
"grad_norm": 3.373734474182129,
|
| 1065 |
+
"learning_rate": 1.1332567489948307e-05,
|
| 1066 |
+
"loss": 0.4548,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.8730614589316484,
|
| 1071 |
+
"grad_norm": 4.657841205596924,
|
| 1072 |
+
"learning_rate": 1.1275129236071226e-05,
|
| 1073 |
+
"loss": 0.36,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.8788052843193567,
|
| 1078 |
+
"grad_norm": 8.218329429626465,
|
| 1079 |
+
"learning_rate": 1.1217690982194142e-05,
|
| 1080 |
+
"loss": 0.4297,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.8845491097070649,
|
| 1085 |
+
"grad_norm": 7.709052562713623,
|
| 1086 |
+
"learning_rate": 1.116025272831706e-05,
|
| 1087 |
+
"loss": 0.3766,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.8902929350947731,
|
| 1092 |
+
"grad_norm": 6.875143527984619,
|
| 1093 |
+
"learning_rate": 1.1102814474439977e-05,
|
| 1094 |
+
"loss": 0.4106,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.8960367604824814,
|
| 1099 |
+
"grad_norm": 5.460892200469971,
|
| 1100 |
+
"learning_rate": 1.1045376220562896e-05,
|
| 1101 |
+
"loss": 0.3016,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.9017805858701895,
|
| 1106 |
+
"grad_norm": 3.4830429553985596,
|
| 1107 |
+
"learning_rate": 1.0987937966685814e-05,
|
| 1108 |
+
"loss": 0.371,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.9075244112578977,
|
| 1113 |
+
"grad_norm": 8.233579635620117,
|
| 1114 |
+
"learning_rate": 1.093049971280873e-05,
|
| 1115 |
+
"loss": 0.4591,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.913268236645606,
|
| 1120 |
+
"grad_norm": 7.001081466674805,
|
| 1121 |
+
"learning_rate": 1.0873061458931649e-05,
|
| 1122 |
+
"loss": 0.3856,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.9190120620333142,
|
| 1127 |
+
"grad_norm": 7.473963260650635,
|
| 1128 |
+
"learning_rate": 1.0815623205054565e-05,
|
| 1129 |
+
"loss": 0.4937,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.9247558874210224,
|
| 1134 |
+
"grad_norm": 4.046863079071045,
|
| 1135 |
+
"learning_rate": 1.0758184951177485e-05,
|
| 1136 |
+
"loss": 0.4141,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.9304997128087306,
|
| 1141 |
+
"grad_norm": 5.425885200500488,
|
| 1142 |
+
"learning_rate": 1.0700746697300404e-05,
|
| 1143 |
+
"loss": 0.3545,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.9362435381964388,
|
| 1148 |
+
"grad_norm": 5.255967140197754,
|
| 1149 |
+
"learning_rate": 1.0643308443423322e-05,
|
| 1150 |
+
"loss": 0.3881,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.941987363584147,
|
| 1155 |
+
"grad_norm": 7.365703105926514,
|
| 1156 |
+
"learning_rate": 1.0585870189546239e-05,
|
| 1157 |
+
"loss": 0.3648,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.9477311889718553,
|
| 1162 |
+
"grad_norm": 5.149658679962158,
|
| 1163 |
+
"learning_rate": 1.0528431935669157e-05,
|
| 1164 |
+
"loss": 0.2952,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.9534750143595635,
|
| 1169 |
+
"grad_norm": 5.5968194007873535,
|
| 1170 |
+
"learning_rate": 1.0470993681792074e-05,
|
| 1171 |
+
"loss": 0.3604,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.9592188397472717,
|
| 1176 |
+
"grad_norm": 6.176368713378906,
|
| 1177 |
+
"learning_rate": 1.0413555427914992e-05,
|
| 1178 |
+
"loss": 0.3584,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.9649626651349799,
|
| 1183 |
+
"grad_norm": 5.876338958740234,
|
| 1184 |
+
"learning_rate": 1.035611717403791e-05,
|
| 1185 |
+
"loss": 0.4076,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.9707064905226881,
|
| 1190 |
+
"grad_norm": 11.697908401489258,
|
| 1191 |
+
"learning_rate": 1.0298678920160827e-05,
|
| 1192 |
+
"loss": 0.3648,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.9764503159103963,
|
| 1197 |
+
"grad_norm": 7.04163122177124,
|
| 1198 |
+
"learning_rate": 1.0241240666283745e-05,
|
| 1199 |
+
"loss": 0.349,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.9821941412981046,
|
| 1204 |
+
"grad_norm": 6.7707133293151855,
|
| 1205 |
+
"learning_rate": 1.0183802412406662e-05,
|
| 1206 |
+
"loss": 0.3888,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.9879379666858128,
|
| 1211 |
+
"grad_norm": 3.8108270168304443,
|
| 1212 |
+
"learning_rate": 1.0126364158529582e-05,
|
| 1213 |
+
"loss": 0.3753,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.9936817920735209,
|
| 1218 |
+
"grad_norm": 11.013320922851562,
|
| 1219 |
+
"learning_rate": 1.00689259046525e-05,
|
| 1220 |
+
"loss": 0.3977,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.9994256174612292,
|
| 1225 |
+
"grad_norm": 4.042791843414307,
|
| 1226 |
+
"learning_rate": 1.0011487650775419e-05,
|
| 1227 |
+
"loss": 0.3724,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.0051694428489375,
|
| 1232 |
+
"grad_norm": 6.258309841156006,
|
| 1233 |
+
"learning_rate": 9.954049396898335e-06,
|
| 1234 |
+
"loss": 0.3591,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.0109132682366455,
|
| 1239 |
+
"grad_norm": 7.884782314300537,
|
| 1240 |
+
"learning_rate": 9.896611143021253e-06,
|
| 1241 |
+
"loss": 0.4114,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.0166570936243537,
|
| 1246 |
+
"grad_norm": 4.9663567543029785,
|
| 1247 |
+
"learning_rate": 9.83917288914417e-06,
|
| 1248 |
+
"loss": 0.4462,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.022400919012062,
|
| 1253 |
+
"grad_norm": 7.046320915222168,
|
| 1254 |
+
"learning_rate": 9.781734635267088e-06,
|
| 1255 |
+
"loss": 0.3386,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.0281447443997702,
|
| 1260 |
+
"grad_norm": 6.846945762634277,
|
| 1261 |
+
"learning_rate": 9.724296381390007e-06,
|
| 1262 |
+
"loss": 0.3181,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.0338885697874785,
|
| 1267 |
+
"grad_norm": 5.925526142120361,
|
| 1268 |
+
"learning_rate": 9.666858127512925e-06,
|
| 1269 |
+
"loss": 0.3357,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.0396323951751867,
|
| 1274 |
+
"grad_norm": 14.302725791931152,
|
| 1275 |
+
"learning_rate": 9.609419873635842e-06,
|
| 1276 |
+
"loss": 0.2859,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.045376220562895,
|
| 1281 |
+
"grad_norm": 8.27291488647461,
|
| 1282 |
+
"learning_rate": 9.55198161975876e-06,
|
| 1283 |
+
"loss": 0.3688,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.0511200459506032,
|
| 1288 |
+
"grad_norm": 5.266950607299805,
|
| 1289 |
+
"learning_rate": 9.494543365881678e-06,
|
| 1290 |
+
"loss": 0.3106,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.0568638713383114,
|
| 1295 |
+
"grad_norm": 11.347005844116211,
|
| 1296 |
+
"learning_rate": 9.437105112004595e-06,
|
| 1297 |
+
"loss": 0.3389,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.0626076967260196,
|
| 1302 |
+
"grad_norm": 4.7072906494140625,
|
| 1303 |
+
"learning_rate": 9.379666858127515e-06,
|
| 1304 |
+
"loss": 0.3519,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.0683515221137276,
|
| 1309 |
+
"grad_norm": 4.05309534072876,
|
| 1310 |
+
"learning_rate": 9.322228604250432e-06,
|
| 1311 |
+
"loss": 0.3006,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.0740953475014359,
|
| 1316 |
+
"grad_norm": 5.578520774841309,
|
| 1317 |
+
"learning_rate": 9.26479035037335e-06,
|
| 1318 |
+
"loss": 0.3645,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.079839172889144,
|
| 1323 |
+
"grad_norm": 7.405791282653809,
|
| 1324 |
+
"learning_rate": 9.207352096496266e-06,
|
| 1325 |
+
"loss": 0.3104,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.0855829982768523,
|
| 1330 |
+
"grad_norm": 9.269173622131348,
|
| 1331 |
+
"learning_rate": 9.149913842619185e-06,
|
| 1332 |
+
"loss": 0.3576,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.0913268236645606,
|
| 1337 |
+
"grad_norm": 5.276297569274902,
|
| 1338 |
+
"learning_rate": 9.092475588742103e-06,
|
| 1339 |
+
"loss": 0.3354,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.0970706490522688,
|
| 1344 |
+
"grad_norm": 8.320406913757324,
|
| 1345 |
+
"learning_rate": 9.035037334865021e-06,
|
| 1346 |
+
"loss": 0.3232,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.102814474439977,
|
| 1351 |
+
"grad_norm": 6.023215293884277,
|
| 1352 |
+
"learning_rate": 8.977599080987938e-06,
|
| 1353 |
+
"loss": 0.3427,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.1085582998276853,
|
| 1358 |
+
"grad_norm": 8.178590774536133,
|
| 1359 |
+
"learning_rate": 8.920160827110856e-06,
|
| 1360 |
+
"loss": 0.3103,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.1143021252153935,
|
| 1365 |
+
"grad_norm": 6.056619644165039,
|
| 1366 |
+
"learning_rate": 8.862722573233775e-06,
|
| 1367 |
+
"loss": 0.3848,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.1200459506031017,
|
| 1372 |
+
"grad_norm": 6.109485626220703,
|
| 1373 |
+
"learning_rate": 8.805284319356693e-06,
|
| 1374 |
+
"loss": 0.3102,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.12578977599081,
|
| 1379 |
+
"grad_norm": 6.949984550476074,
|
| 1380 |
+
"learning_rate": 8.747846065479611e-06,
|
| 1381 |
+
"loss": 0.3013,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.1315336013785182,
|
| 1386 |
+
"grad_norm": 4.320880889892578,
|
| 1387 |
+
"learning_rate": 8.690407811602528e-06,
|
| 1388 |
+
"loss": 0.3169,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.1372774267662262,
|
| 1393 |
+
"grad_norm": 9.62964916229248,
|
| 1394 |
+
"learning_rate": 8.632969557725446e-06,
|
| 1395 |
+
"loss": 0.3577,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.1430212521539345,
|
| 1400 |
+
"grad_norm": 7.1865105628967285,
|
| 1401 |
+
"learning_rate": 8.575531303848363e-06,
|
| 1402 |
+
"loss": 0.4245,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.1487650775416427,
|
| 1407 |
+
"grad_norm": 11.42944622039795,
|
| 1408 |
+
"learning_rate": 8.518093049971281e-06,
|
| 1409 |
+
"loss": 0.3421,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.154508902929351,
|
| 1414 |
+
"grad_norm": 10.365814208984375,
|
| 1415 |
+
"learning_rate": 8.4606547960942e-06,
|
| 1416 |
+
"loss": 0.2971,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.1602527283170592,
|
| 1421 |
+
"grad_norm": 4.546888828277588,
|
| 1422 |
+
"learning_rate": 8.403216542217118e-06,
|
| 1423 |
+
"loss": 0.3762,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.1659965537047674,
|
| 1428 |
+
"grad_norm": 9.672823905944824,
|
| 1429 |
+
"learning_rate": 8.345778288340034e-06,
|
| 1430 |
+
"loss": 0.3411,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.1717403790924756,
|
| 1435 |
+
"grad_norm": 4.738915920257568,
|
| 1436 |
+
"learning_rate": 8.288340034462953e-06,
|
| 1437 |
+
"loss": 0.3453,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.1774842044801839,
|
| 1442 |
+
"grad_norm": 10.187810897827148,
|
| 1443 |
+
"learning_rate": 8.230901780585871e-06,
|
| 1444 |
+
"loss": 0.3351,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 1.183228029867892,
|
| 1449 |
+
"grad_norm": 6.290671348571777,
|
| 1450 |
+
"learning_rate": 8.17346352670879e-06,
|
| 1451 |
+
"loss": 0.3286,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.1889718552556001,
|
| 1456 |
+
"grad_norm": 9.14261531829834,
|
| 1457 |
+
"learning_rate": 8.116025272831708e-06,
|
| 1458 |
+
"loss": 0.2878,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.1947156806433084,
|
| 1463 |
+
"grad_norm": 7.814758777618408,
|
| 1464 |
+
"learning_rate": 8.058587018954624e-06,
|
| 1465 |
+
"loss": 0.3469,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.2004595060310166,
|
| 1470 |
+
"grad_norm": 10.085731506347656,
|
| 1471 |
+
"learning_rate": 8.001148765077543e-06,
|
| 1472 |
+
"loss": 0.3025,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.2062033314187248,
|
| 1477 |
+
"grad_norm": 10.734376907348633,
|
| 1478 |
+
"learning_rate": 7.94371051120046e-06,
|
| 1479 |
+
"loss": 0.2442,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.211947156806433,
|
| 1484 |
+
"grad_norm": 13.61286735534668,
|
| 1485 |
+
"learning_rate": 7.88627225732338e-06,
|
| 1486 |
+
"loss": 0.3063,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.2176909821941413,
|
| 1491 |
+
"grad_norm": 8.572850227355957,
|
| 1492 |
+
"learning_rate": 7.828834003446296e-06,
|
| 1493 |
+
"loss": 0.3863,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 1.2234348075818495,
|
| 1498 |
+
"grad_norm": 6.247170448303223,
|
| 1499 |
+
"learning_rate": 7.771395749569214e-06,
|
| 1500 |
+
"loss": 0.3214,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 1.2291786329695578,
|
| 1505 |
+
"grad_norm": 7.438636779785156,
|
| 1506 |
+
"learning_rate": 7.71395749569213e-06,
|
| 1507 |
+
"loss": 0.381,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 1.234922458357266,
|
| 1512 |
+
"grad_norm": 6.11846399307251,
|
| 1513 |
+
"learning_rate": 7.656519241815049e-06,
|
| 1514 |
+
"loss": 0.3556,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 1.2406662837449742,
|
| 1519 |
+
"grad_norm": 10.697092056274414,
|
| 1520 |
+
"learning_rate": 7.5990809879379666e-06,
|
| 1521 |
+
"loss": 0.3767,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 1.2464101091326825,
|
| 1526 |
+
"grad_norm": 5.3118205070495605,
|
| 1527 |
+
"learning_rate": 7.541642734060886e-06,
|
| 1528 |
+
"loss": 0.3112,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 1.2521539345203907,
|
| 1533 |
+
"grad_norm": 5.907925128936768,
|
| 1534 |
+
"learning_rate": 7.484204480183803e-06,
|
| 1535 |
+
"loss": 0.2833,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 1.2578977599080987,
|
| 1540 |
+
"grad_norm": 7.271302223205566,
|
| 1541 |
+
"learning_rate": 7.426766226306721e-06,
|
| 1542 |
+
"loss": 0.1935,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 1.263641585295807,
|
| 1547 |
+
"grad_norm": 12.389423370361328,
|
| 1548 |
+
"learning_rate": 7.369327972429638e-06,
|
| 1549 |
+
"loss": 0.3946,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.2693854106835152,
|
| 1554 |
+
"grad_norm": 9.09422492980957,
|
| 1555 |
+
"learning_rate": 7.3118897185525564e-06,
|
| 1556 |
+
"loss": 0.3191,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.2751292360712234,
|
| 1561 |
+
"grad_norm": 8.75156307220459,
|
| 1562 |
+
"learning_rate": 7.254451464675475e-06,
|
| 1563 |
+
"loss": 0.4077,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.2808730614589316,
|
| 1568 |
+
"grad_norm": 7.306863784790039,
|
| 1569 |
+
"learning_rate": 7.197013210798392e-06,
|
| 1570 |
+
"loss": 0.3295,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.2866168868466399,
|
| 1575 |
+
"grad_norm": 9.715473175048828,
|
| 1576 |
+
"learning_rate": 7.1395749569213105e-06,
|
| 1577 |
+
"loss": 0.4163,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 1.2923607122343481,
|
| 1582 |
+
"grad_norm": 6.315252780914307,
|
| 1583 |
+
"learning_rate": 7.082136703044228e-06,
|
| 1584 |
+
"loss": 0.3817,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.2981045376220564,
|
| 1589 |
+
"grad_norm": 8.821859359741211,
|
| 1590 |
+
"learning_rate": 7.0246984491671455e-06,
|
| 1591 |
+
"loss": 0.3265,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.3038483630097644,
|
| 1596 |
+
"grad_norm": 6.838233947753906,
|
| 1597 |
+
"learning_rate": 6.967260195290065e-06,
|
| 1598 |
+
"loss": 0.2966,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 1.3095921883974726,
|
| 1603 |
+
"grad_norm": 9.925073623657227,
|
| 1604 |
+
"learning_rate": 6.909821941412982e-06,
|
| 1605 |
+
"loss": 0.4484,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.3153360137851808,
|
| 1610 |
+
"grad_norm": 5.026411056518555,
|
| 1611 |
+
"learning_rate": 6.8523836875358996e-06,
|
| 1612 |
+
"loss": 0.4647,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 1.321079839172889,
|
| 1617 |
+
"grad_norm": 4.3956732749938965,
|
| 1618 |
+
"learning_rate": 6.794945433658817e-06,
|
| 1619 |
+
"loss": 0.2968,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 1.3268236645605973,
|
| 1624 |
+
"grad_norm": 6.904971599578857,
|
| 1625 |
+
"learning_rate": 6.7375071797817345e-06,
|
| 1626 |
+
"loss": 0.2888,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 1.3325674899483055,
|
| 1631 |
+
"grad_norm": 3.1684281826019287,
|
| 1632 |
+
"learning_rate": 6.680068925904653e-06,
|
| 1633 |
+
"loss": 0.2975,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 1.3383113153360138,
|
| 1638 |
+
"grad_norm": 7.3333420753479,
|
| 1639 |
+
"learning_rate": 6.622630672027571e-06,
|
| 1640 |
+
"loss": 0.3911,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 1.344055140723722,
|
| 1645 |
+
"grad_norm": 7.822445392608643,
|
| 1646 |
+
"learning_rate": 6.565192418150489e-06,
|
| 1647 |
+
"loss": 0.3199,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 1.3497989661114302,
|
| 1652 |
+
"grad_norm": 9.02872371673584,
|
| 1653 |
+
"learning_rate": 6.507754164273407e-06,
|
| 1654 |
+
"loss": 0.4698,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 1.3555427914991385,
|
| 1659 |
+
"grad_norm": 3.9332520961761475,
|
| 1660 |
+
"learning_rate": 6.450315910396324e-06,
|
| 1661 |
+
"loss": 0.3651,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 1.3612866168868467,
|
| 1666 |
+
"grad_norm": 7.590347766876221,
|
| 1667 |
+
"learning_rate": 6.392877656519242e-06,
|
| 1668 |
+
"loss": 0.3121,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 1.367030442274555,
|
| 1673 |
+
"grad_norm": 8.964584350585938,
|
| 1674 |
+
"learning_rate": 6.335439402642161e-06,
|
| 1675 |
+
"loss": 0.271,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 1.3727742676622632,
|
| 1680 |
+
"grad_norm": 8.058918952941895,
|
| 1681 |
+
"learning_rate": 6.2780011487650785e-06,
|
| 1682 |
+
"loss": 0.3906,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 1.3785180930499714,
|
| 1687 |
+
"grad_norm": 6.742099761962891,
|
| 1688 |
+
"learning_rate": 6.220562894887996e-06,
|
| 1689 |
+
"loss": 0.3072,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 1.3842619184376794,
|
| 1694 |
+
"grad_norm": 5.961569309234619,
|
| 1695 |
+
"learning_rate": 6.1631246410109134e-06,
|
| 1696 |
+
"loss": 0.3673,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 1.3900057438253877,
|
| 1701 |
+
"grad_norm": 9.705893516540527,
|
| 1702 |
+
"learning_rate": 6.105686387133831e-06,
|
| 1703 |
+
"loss": 0.3544,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 1.395749569213096,
|
| 1708 |
+
"grad_norm": 4.435375690460205,
|
| 1709 |
+
"learning_rate": 6.04824813325675e-06,
|
| 1710 |
+
"loss": 0.2372,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 1.4014933946008041,
|
| 1715 |
+
"grad_norm": 5.375720977783203,
|
| 1716 |
+
"learning_rate": 5.9908098793796675e-06,
|
| 1717 |
+
"loss": 0.2264,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 1.4072372199885124,
|
| 1722 |
+
"grad_norm": 5.602358818054199,
|
| 1723 |
+
"learning_rate": 5.933371625502585e-06,
|
| 1724 |
+
"loss": 0.3449,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 1.4129810453762206,
|
| 1729 |
+
"grad_norm": 10.811373710632324,
|
| 1730 |
+
"learning_rate": 5.875933371625503e-06,
|
| 1731 |
+
"loss": 0.3663,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 1.4187248707639288,
|
| 1736 |
+
"grad_norm": 10.196518898010254,
|
| 1737 |
+
"learning_rate": 5.818495117748421e-06,
|
| 1738 |
+
"loss": 0.3546,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 1.424468696151637,
|
| 1743 |
+
"grad_norm": 10.06306266784668,
|
| 1744 |
+
"learning_rate": 5.761056863871339e-06,
|
| 1745 |
+
"loss": 0.3282,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 1.430212521539345,
|
| 1750 |
+
"grad_norm": 4.978325843811035,
|
| 1751 |
+
"learning_rate": 5.703618609994257e-06,
|
| 1752 |
+
"loss": 0.2961,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 1.4359563469270533,
|
| 1757 |
+
"grad_norm": 10.731146812438965,
|
| 1758 |
+
"learning_rate": 5.646180356117175e-06,
|
| 1759 |
+
"loss": 0.3349,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 1.4417001723147616,
|
| 1764 |
+
"grad_norm": 8.913891792297363,
|
| 1765 |
+
"learning_rate": 5.588742102240092e-06,
|
| 1766 |
+
"loss": 0.3131,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 1.4474439977024698,
|
| 1771 |
+
"grad_norm": 5.1745195388793945,
|
| 1772 |
+
"learning_rate": 5.53130384836301e-06,
|
| 1773 |
+
"loss": 0.3842,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 1.453187823090178,
|
| 1778 |
+
"grad_norm": 8.361491203308105,
|
| 1779 |
+
"learning_rate": 5.473865594485927e-06,
|
| 1780 |
+
"loss": 0.3598,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 1.4589316484778863,
|
| 1785 |
+
"grad_norm": 6.487078666687012,
|
| 1786 |
+
"learning_rate": 5.4164273406088464e-06,
|
| 1787 |
+
"loss": 0.3244,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 1.4646754738655945,
|
| 1792 |
+
"grad_norm": 4.129726409912109,
|
| 1793 |
+
"learning_rate": 5.358989086731764e-06,
|
| 1794 |
+
"loss": 0.3152,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 1.4704192992533027,
|
| 1799 |
+
"grad_norm": 9.363592147827148,
|
| 1800 |
+
"learning_rate": 5.301550832854681e-06,
|
| 1801 |
+
"loss": 0.3321,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 1.476163124641011,
|
| 1806 |
+
"grad_norm": 6.334773063659668,
|
| 1807 |
+
"learning_rate": 5.2441125789776e-06,
|
| 1808 |
+
"loss": 0.3312,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 1.4819069500287192,
|
| 1813 |
+
"grad_norm": 7.404930114746094,
|
| 1814 |
+
"learning_rate": 5.186674325100517e-06,
|
| 1815 |
+
"loss": 0.3739,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 1.4876507754164274,
|
| 1820 |
+
"grad_norm": 7.487016201019287,
|
| 1821 |
+
"learning_rate": 5.1292360712234355e-06,
|
| 1822 |
+
"loss": 0.2865,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 1.4933946008041357,
|
| 1827 |
+
"grad_norm": 13.322307586669922,
|
| 1828 |
+
"learning_rate": 5.071797817346353e-06,
|
| 1829 |
+
"loss": 0.3444,
|
| 1830 |
+
"step": 2600
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 1.499138426191844,
|
| 1834 |
+
"grad_norm": 9.053878784179688,
|
| 1835 |
+
"learning_rate": 5.014359563469271e-06,
|
| 1836 |
+
"loss": 0.3799,
|
| 1837 |
+
"step": 2610
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 1.5048822515795521,
|
| 1841 |
+
"grad_norm": 4.018943786621094,
|
| 1842 |
+
"learning_rate": 4.956921309592189e-06,
|
| 1843 |
+
"loss": 0.2567,
|
| 1844 |
+
"step": 2620
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 1.5106260769672601,
|
| 1848 |
+
"grad_norm": 2.2457354068756104,
|
| 1849 |
+
"learning_rate": 4.899483055715107e-06,
|
| 1850 |
+
"loss": 0.2978,
|
| 1851 |
+
"step": 2630
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 1.5163699023549684,
|
| 1855 |
+
"grad_norm": 4.894889831542969,
|
| 1856 |
+
"learning_rate": 4.8420448018380245e-06,
|
| 1857 |
+
"loss": 0.2861,
|
| 1858 |
+
"step": 2640
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 1.5221137277426766,
|
| 1862 |
+
"grad_norm": 6.843629360198975,
|
| 1863 |
+
"learning_rate": 4.784606547960942e-06,
|
| 1864 |
+
"loss": 0.3254,
|
| 1865 |
+
"step": 2650
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 1.5278575531303848,
|
| 1869 |
+
"grad_norm": 7.173573970794678,
|
| 1870 |
+
"learning_rate": 4.72716829408386e-06,
|
| 1871 |
+
"loss": 0.3481,
|
| 1872 |
+
"step": 2660
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 1.533601378518093,
|
| 1876 |
+
"grad_norm": 12.328543663024902,
|
| 1877 |
+
"learning_rate": 4.669730040206778e-06,
|
| 1878 |
+
"loss": 0.3124,
|
| 1879 |
+
"step": 2670
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 1.5393452039058013,
|
| 1883 |
+
"grad_norm": 9.337592124938965,
|
| 1884 |
+
"learning_rate": 4.612291786329696e-06,
|
| 1885 |
+
"loss": 0.3005,
|
| 1886 |
+
"step": 2680
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 1.5450890292935093,
|
| 1890 |
+
"grad_norm": 5.477969646453857,
|
| 1891 |
+
"learning_rate": 4.5548535324526135e-06,
|
| 1892 |
+
"loss": 0.3457,
|
| 1893 |
+
"step": 2690
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 1.5508328546812176,
|
| 1897 |
+
"grad_norm": 5.083920955657959,
|
| 1898 |
+
"learning_rate": 4.497415278575532e-06,
|
| 1899 |
+
"loss": 0.2858,
|
| 1900 |
+
"step": 2700
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 1.5565766800689258,
|
| 1904 |
+
"grad_norm": 6.250855445861816,
|
| 1905 |
+
"learning_rate": 4.439977024698449e-06,
|
| 1906 |
+
"loss": 0.2673,
|
| 1907 |
+
"step": 2710
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 1.562320505456634,
|
| 1911 |
+
"grad_norm": 6.169952392578125,
|
| 1912 |
+
"learning_rate": 4.382538770821368e-06,
|
| 1913 |
+
"loss": 0.2971,
|
| 1914 |
+
"step": 2720
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 1.5680643308443423,
|
| 1918 |
+
"grad_norm": 8.261754989624023,
|
| 1919 |
+
"learning_rate": 4.325100516944285e-06,
|
| 1920 |
+
"loss": 0.3009,
|
| 1921 |
+
"step": 2730
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 1.5738081562320505,
|
| 1925 |
+
"grad_norm": 8.477384567260742,
|
| 1926 |
+
"learning_rate": 4.267662263067203e-06,
|
| 1927 |
+
"loss": 0.3017,
|
| 1928 |
+
"step": 2740
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 1.5795519816197587,
|
| 1932 |
+
"grad_norm": 8.52374267578125,
|
| 1933 |
+
"learning_rate": 4.210224009190121e-06,
|
| 1934 |
+
"loss": 0.4511,
|
| 1935 |
+
"step": 2750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 1.585295807007467,
|
| 1939 |
+
"grad_norm": 8.646858215332031,
|
| 1940 |
+
"learning_rate": 4.152785755313039e-06,
|
| 1941 |
+
"loss": 0.2699,
|
| 1942 |
+
"step": 2760
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 1.5910396323951752,
|
| 1946 |
+
"grad_norm": 7.500174522399902,
|
| 1947 |
+
"learning_rate": 4.095347501435957e-06,
|
| 1948 |
+
"loss": 0.2743,
|
| 1949 |
+
"step": 2770
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 1.5967834577828834,
|
| 1953 |
+
"grad_norm": 5.454465389251709,
|
| 1954 |
+
"learning_rate": 4.037909247558874e-06,
|
| 1955 |
+
"loss": 0.3023,
|
| 1956 |
+
"step": 2780
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 1.6025272831705917,
|
| 1960 |
+
"grad_norm": 2.6998019218444824,
|
| 1961 |
+
"learning_rate": 3.9804709936817925e-06,
|
| 1962 |
+
"loss": 0.3941,
|
| 1963 |
+
"step": 2790
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 1.6082711085583,
|
| 1967 |
+
"grad_norm": 4.594570159912109,
|
| 1968 |
+
"learning_rate": 3.92303273980471e-06,
|
| 1969 |
+
"loss": 0.252,
|
| 1970 |
+
"step": 2800
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 1.6140149339460081,
|
| 1974 |
+
"grad_norm": 5.87538480758667,
|
| 1975 |
+
"learning_rate": 3.865594485927628e-06,
|
| 1976 |
+
"loss": 0.3093,
|
| 1977 |
+
"step": 2810
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 1.6197587593337164,
|
| 1981 |
+
"grad_norm": 5.358250617980957,
|
| 1982 |
+
"learning_rate": 3.808156232050546e-06,
|
| 1983 |
+
"loss": 0.3674,
|
| 1984 |
+
"step": 2820
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 1.6255025847214246,
|
| 1988 |
+
"grad_norm": 4.871222972869873,
|
| 1989 |
+
"learning_rate": 3.7507179781734636e-06,
|
| 1990 |
+
"loss": 0.2407,
|
| 1991 |
+
"step": 2830
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 1.6312464101091326,
|
| 1995 |
+
"grad_norm": 6.6836838722229,
|
| 1996 |
+
"learning_rate": 3.693279724296382e-06,
|
| 1997 |
+
"loss": 0.2978,
|
| 1998 |
+
"step": 2840
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 1.6369902354968409,
|
| 2002 |
+
"grad_norm": 7.67780065536499,
|
| 2003 |
+
"learning_rate": 3.6358414704192994e-06,
|
| 2004 |
+
"loss": 0.3655,
|
| 2005 |
+
"step": 2850
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 1.642734060884549,
|
| 2009 |
+
"grad_norm": 6.890137672424316,
|
| 2010 |
+
"learning_rate": 3.5784032165422173e-06,
|
| 2011 |
+
"loss": 0.3618,
|
| 2012 |
+
"step": 2860
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 1.6484778862722573,
|
| 2016 |
+
"grad_norm": 7.769665241241455,
|
| 2017 |
+
"learning_rate": 3.5209649626651356e-06,
|
| 2018 |
+
"loss": 0.2703,
|
| 2019 |
+
"step": 2870
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 1.6542217116599656,
|
| 2023 |
+
"grad_norm": 4.888299465179443,
|
| 2024 |
+
"learning_rate": 3.463526708788053e-06,
|
| 2025 |
+
"loss": 0.3257,
|
| 2026 |
+
"step": 2880
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 1.6599655370476738,
|
| 2030 |
+
"grad_norm": 4.726266384124756,
|
| 2031 |
+
"learning_rate": 3.406088454910971e-06,
|
| 2032 |
+
"loss": 0.3573,
|
| 2033 |
+
"step": 2890
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 1.6657093624353818,
|
| 2037 |
+
"grad_norm": 6.836297035217285,
|
| 2038 |
+
"learning_rate": 3.348650201033889e-06,
|
| 2039 |
+
"loss": 0.3325,
|
| 2040 |
+
"step": 2900
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 1.67145318782309,
|
| 2044 |
+
"grad_norm": 6.571508884429932,
|
| 2045 |
+
"learning_rate": 3.2912119471568067e-06,
|
| 2046 |
+
"loss": 0.292,
|
| 2047 |
+
"step": 2910
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 1.6771970132107983,
|
| 2051 |
+
"grad_norm": 9.843769073486328,
|
| 2052 |
+
"learning_rate": 3.2337736932797246e-06,
|
| 2053 |
+
"loss": 0.4027,
|
| 2054 |
+
"step": 2920
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 1.6829408385985065,
|
| 2058 |
+
"grad_norm": 6.089911937713623,
|
| 2059 |
+
"learning_rate": 3.1763354394026425e-06,
|
| 2060 |
+
"loss": 0.3627,
|
| 2061 |
+
"step": 2930
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 1.6886846639862148,
|
| 2065 |
+
"grad_norm": 6.846927165985107,
|
| 2066 |
+
"learning_rate": 3.11889718552556e-06,
|
| 2067 |
+
"loss": 0.3541,
|
| 2068 |
+
"step": 2940
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 1.694428489373923,
|
| 2072 |
+
"grad_norm": 10.23181438446045,
|
| 2073 |
+
"learning_rate": 3.0614589316484783e-06,
|
| 2074 |
+
"loss": 0.3418,
|
| 2075 |
+
"step": 2950
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 1.7001723147616312,
|
| 2079 |
+
"grad_norm": 5.403523921966553,
|
| 2080 |
+
"learning_rate": 3.0040206777713958e-06,
|
| 2081 |
+
"loss": 0.3498,
|
| 2082 |
+
"step": 2960
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 1.7059161401493395,
|
| 2086 |
+
"grad_norm": 8.252917289733887,
|
| 2087 |
+
"learning_rate": 2.9465824238943137e-06,
|
| 2088 |
+
"loss": 0.3401,
|
| 2089 |
+
"step": 2970
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 1.7116599655370477,
|
| 2093 |
+
"grad_norm": 7.06523323059082,
|
| 2094 |
+
"learning_rate": 2.889144170017232e-06,
|
| 2095 |
+
"loss": 0.2331,
|
| 2096 |
+
"step": 2980
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 1.717403790924756,
|
| 2100 |
+
"grad_norm": 7.739984035491943,
|
| 2101 |
+
"learning_rate": 2.8317059161401494e-06,
|
| 2102 |
+
"loss": 0.3678,
|
| 2103 |
+
"step": 2990
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 1.7231476163124642,
|
| 2107 |
+
"grad_norm": 4.021157741546631,
|
| 2108 |
+
"learning_rate": 2.7742676622630677e-06,
|
| 2109 |
+
"loss": 0.3537,
|
| 2110 |
+
"step": 3000
|
| 2111 |
+
}
|
| 2112 |
+
],
|
| 2113 |
+
"logging_steps": 10,
|
| 2114 |
+
"max_steps": 3482,
|
| 2115 |
+
"num_input_tokens_seen": 0,
|
| 2116 |
+
"num_train_epochs": 2,
|
| 2117 |
+
"save_steps": 500,
|
| 2118 |
+
"stateful_callbacks": {
|
| 2119 |
+
"TrainerControl": {
|
| 2120 |
+
"args": {
|
| 2121 |
+
"should_epoch_stop": false,
|
| 2122 |
+
"should_evaluate": false,
|
| 2123 |
+
"should_log": false,
|
| 2124 |
+
"should_save": true,
|
| 2125 |
+
"should_training_stop": false
|
| 2126 |
+
},
|
| 2127 |
+
"attributes": {}
|
| 2128 |
+
}
|
| 2129 |
+
},
|
| 2130 |
+
"total_flos": 6356845526704128.0,
|
| 2131 |
+
"train_batch_size": 16,
|
| 2132 |
+
"trial_name": null,
|
| 2133 |
+
"trial_params": null
|
| 2134 |
+
}
|
checkpoint-3000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7e7cb6431ba79d66c2c9c534739c31af31901b4f69b86ef7cd6fa47d58097e4a
|
| 3 |
+
size 5368
|
checkpoint-3000/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-3482/config.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation": "gelu",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"DistilBertForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.1,
|
| 7 |
+
"dim": 768,
|
| 8 |
+
"dropout": 0.3,
|
| 9 |
+
"dtype": "float32",
|
| 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 |
+
"transformers_version": "4.57.1",
|
| 23 |
+
"vocab_size": 30522
|
| 24 |
+
}
|
checkpoint-3482/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:44a9699f5eae1c12be68d7d6783b3db0d0136ec5a126bd85d7435845fcefef10
|
| 3 |
+
size 267832560
|
checkpoint-3482/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b99703846da685dc1ba9b58eb1ffbaa4ed9817271ddc7bd75053cfd6a8a0983b
|
| 3 |
+
size 535727290
|
checkpoint-3482/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e1af46080ed57ad2b99923ba401a4515d7b1bc2bc42a980f6723e8b5616b10fa
|
| 3 |
+
size 14244
|
checkpoint-3482/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9ed4587eaad0c9722ca4447ada6e6b52ba5a765ac9839bc3b828962bebeed812
|
| 3 |
+
size 1064
|
checkpoint-3482/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-3482/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-3482/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-3482/trainer_state.json
ADDED
|
@@ -0,0 +1,2470 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 2.0,
|
| 6 |
+
"eval_steps": 100,
|
| 7 |
+
"global_step": 3482,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.005743825387708214,
|
| 14 |
+
"grad_norm": 1.664604902267456,
|
| 15 |
+
"learning_rate": 1.9948305571510626e-05,
|
| 16 |
+
"loss": 0.6766,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.011487650775416428,
|
| 21 |
+
"grad_norm": 1.8649318218231201,
|
| 22 |
+
"learning_rate": 1.9890867317633546e-05,
|
| 23 |
+
"loss": 0.6443,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.01723147616312464,
|
| 28 |
+
"grad_norm": 2.0676939487457275,
|
| 29 |
+
"learning_rate": 1.9833429063756463e-05,
|
| 30 |
+
"loss": 0.6516,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.022975301550832855,
|
| 35 |
+
"grad_norm": 2.077414035797119,
|
| 36 |
+
"learning_rate": 1.9775990809879383e-05,
|
| 37 |
+
"loss": 0.5468,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.02871912693854107,
|
| 42 |
+
"grad_norm": 2.4310338497161865,
|
| 43 |
+
"learning_rate": 1.97185525560023e-05,
|
| 44 |
+
"loss": 0.6289,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.03446295232624928,
|
| 49 |
+
"grad_norm": 2.605480670928955,
|
| 50 |
+
"learning_rate": 1.9661114302125216e-05,
|
| 51 |
+
"loss": 0.583,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.040206777713957496,
|
| 56 |
+
"grad_norm": 2.7097268104553223,
|
| 57 |
+
"learning_rate": 1.9603676048248136e-05,
|
| 58 |
+
"loss": 0.6029,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.04595060310166571,
|
| 63 |
+
"grad_norm": 4.125054359436035,
|
| 64 |
+
"learning_rate": 1.9546237794371053e-05,
|
| 65 |
+
"loss": 0.5545,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.051694428489373924,
|
| 70 |
+
"grad_norm": 4.592983245849609,
|
| 71 |
+
"learning_rate": 1.948879954049397e-05,
|
| 72 |
+
"loss": 0.5027,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.05743825387708214,
|
| 77 |
+
"grad_norm": 4.799990653991699,
|
| 78 |
+
"learning_rate": 1.9431361286616886e-05,
|
| 79 |
+
"loss": 0.4913,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.06318207926479034,
|
| 84 |
+
"grad_norm": 5.146159648895264,
|
| 85 |
+
"learning_rate": 1.9373923032739806e-05,
|
| 86 |
+
"loss": 0.5419,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.06892590465249857,
|
| 91 |
+
"grad_norm": 6.291711807250977,
|
| 92 |
+
"learning_rate": 1.9316484778862726e-05,
|
| 93 |
+
"loss": 0.478,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.07466973004020677,
|
| 98 |
+
"grad_norm": 4.076691627502441,
|
| 99 |
+
"learning_rate": 1.9259046524985643e-05,
|
| 100 |
+
"loss": 0.489,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.08041355542791499,
|
| 105 |
+
"grad_norm": 5.057150363922119,
|
| 106 |
+
"learning_rate": 1.920160827110856e-05,
|
| 107 |
+
"loss": 0.5435,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.0861573808156232,
|
| 112 |
+
"grad_norm": 3.5335211753845215,
|
| 113 |
+
"learning_rate": 1.914417001723148e-05,
|
| 114 |
+
"loss": 0.4445,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.09190120620333142,
|
| 119 |
+
"grad_norm": 6.209961414337158,
|
| 120 |
+
"learning_rate": 1.9086731763354396e-05,
|
| 121 |
+
"loss": 0.4516,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.09764503159103963,
|
| 126 |
+
"grad_norm": 7.1031341552734375,
|
| 127 |
+
"learning_rate": 1.9029293509477313e-05,
|
| 128 |
+
"loss": 0.576,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.10338885697874785,
|
| 133 |
+
"grad_norm": 4.959975719451904,
|
| 134 |
+
"learning_rate": 1.8971855255600233e-05,
|
| 135 |
+
"loss": 0.4841,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.10913268236645605,
|
| 140 |
+
"grad_norm": 4.904316425323486,
|
| 141 |
+
"learning_rate": 1.891441700172315e-05,
|
| 142 |
+
"loss": 0.5007,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.11487650775416428,
|
| 147 |
+
"grad_norm": 8.833961486816406,
|
| 148 |
+
"learning_rate": 1.8856978747846066e-05,
|
| 149 |
+
"loss": 0.5416,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.12062033314187248,
|
| 154 |
+
"grad_norm": 6.040881156921387,
|
| 155 |
+
"learning_rate": 1.8799540493968982e-05,
|
| 156 |
+
"loss": 0.5162,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.1263641585295807,
|
| 161 |
+
"grad_norm": 7.167252063751221,
|
| 162 |
+
"learning_rate": 1.8742102240091902e-05,
|
| 163 |
+
"loss": 0.4965,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.13210798391728892,
|
| 168 |
+
"grad_norm": 3.9503183364868164,
|
| 169 |
+
"learning_rate": 1.8684663986214822e-05,
|
| 170 |
+
"loss": 0.5129,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.13785180930499713,
|
| 175 |
+
"grad_norm": 4.403937339782715,
|
| 176 |
+
"learning_rate": 1.862722573233774e-05,
|
| 177 |
+
"loss": 0.5192,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.14359563469270534,
|
| 182 |
+
"grad_norm": 6.813930034637451,
|
| 183 |
+
"learning_rate": 1.8569787478460656e-05,
|
| 184 |
+
"loss": 0.4927,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.14933946008041354,
|
| 189 |
+
"grad_norm": 4.355352878570557,
|
| 190 |
+
"learning_rate": 1.8512349224583576e-05,
|
| 191 |
+
"loss": 0.4857,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.15508328546812178,
|
| 196 |
+
"grad_norm": 4.012150287628174,
|
| 197 |
+
"learning_rate": 1.8454910970706492e-05,
|
| 198 |
+
"loss": 0.3763,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.16082711085582999,
|
| 203 |
+
"grad_norm": 7.179994106292725,
|
| 204 |
+
"learning_rate": 1.839747271682941e-05,
|
| 205 |
+
"loss": 0.4625,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.1665709362435382,
|
| 210 |
+
"grad_norm": 3.701215982437134,
|
| 211 |
+
"learning_rate": 1.834003446295233e-05,
|
| 212 |
+
"loss": 0.4954,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.1723147616312464,
|
| 217 |
+
"grad_norm": 5.150369167327881,
|
| 218 |
+
"learning_rate": 1.8282596209075246e-05,
|
| 219 |
+
"loss": 0.4875,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.17805858701895463,
|
| 224 |
+
"grad_norm": 4.432060241699219,
|
| 225 |
+
"learning_rate": 1.8225157955198162e-05,
|
| 226 |
+
"loss": 0.5032,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.18380241240666284,
|
| 231 |
+
"grad_norm": 4.560088634490967,
|
| 232 |
+
"learning_rate": 1.816771970132108e-05,
|
| 233 |
+
"loss": 0.4443,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.18954623779437105,
|
| 238 |
+
"grad_norm": 6.000060558319092,
|
| 239 |
+
"learning_rate": 1.8110281447444e-05,
|
| 240 |
+
"loss": 0.4805,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.19529006318207925,
|
| 245 |
+
"grad_norm": 6.462589740753174,
|
| 246 |
+
"learning_rate": 1.805284319356692e-05,
|
| 247 |
+
"loss": 0.4009,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.2010338885697875,
|
| 252 |
+
"grad_norm": 4.6327104568481445,
|
| 253 |
+
"learning_rate": 1.7995404939689835e-05,
|
| 254 |
+
"loss": 0.5126,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.2067777139574957,
|
| 259 |
+
"grad_norm": 5.640307426452637,
|
| 260 |
+
"learning_rate": 1.7937966685812752e-05,
|
| 261 |
+
"loss": 0.4308,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.2125215393452039,
|
| 266 |
+
"grad_norm": 6.553674697875977,
|
| 267 |
+
"learning_rate": 1.7880528431935672e-05,
|
| 268 |
+
"loss": 0.4686,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.2182653647329121,
|
| 273 |
+
"grad_norm": 6.482476234436035,
|
| 274 |
+
"learning_rate": 1.782309017805859e-05,
|
| 275 |
+
"loss": 0.4214,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.22400919012062034,
|
| 280 |
+
"grad_norm": 6.897400379180908,
|
| 281 |
+
"learning_rate": 1.7765651924181505e-05,
|
| 282 |
+
"loss": 0.4827,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.22975301550832855,
|
| 287 |
+
"grad_norm": 3.935074806213379,
|
| 288 |
+
"learning_rate": 1.7708213670304425e-05,
|
| 289 |
+
"loss": 0.4212,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.23549684089603676,
|
| 294 |
+
"grad_norm": 5.766780853271484,
|
| 295 |
+
"learning_rate": 1.7650775416427342e-05,
|
| 296 |
+
"loss": 0.4421,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.24124066628374496,
|
| 301 |
+
"grad_norm": 8.834909439086914,
|
| 302 |
+
"learning_rate": 1.759333716255026e-05,
|
| 303 |
+
"loss": 0.4676,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.2469844916714532,
|
| 308 |
+
"grad_norm": 4.853402614593506,
|
| 309 |
+
"learning_rate": 1.7535898908673175e-05,
|
| 310 |
+
"loss": 0.4888,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.2527283170591614,
|
| 315 |
+
"grad_norm": 5.489238739013672,
|
| 316 |
+
"learning_rate": 1.7478460654796095e-05,
|
| 317 |
+
"loss": 0.3763,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.2584721424468696,
|
| 322 |
+
"grad_norm": 3.805001974105835,
|
| 323 |
+
"learning_rate": 1.7421022400919015e-05,
|
| 324 |
+
"loss": 0.5266,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.26421596783457785,
|
| 329 |
+
"grad_norm": 5.9350080490112305,
|
| 330 |
+
"learning_rate": 1.7363584147041932e-05,
|
| 331 |
+
"loss": 0.4053,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.269959793222286,
|
| 336 |
+
"grad_norm": 8.31826400756836,
|
| 337 |
+
"learning_rate": 1.730614589316485e-05,
|
| 338 |
+
"loss": 0.4775,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.27570361860999426,
|
| 343 |
+
"grad_norm": 5.709866046905518,
|
| 344 |
+
"learning_rate": 1.724870763928777e-05,
|
| 345 |
+
"loss": 0.3681,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.2814474439977025,
|
| 350 |
+
"grad_norm": 2.6657683849334717,
|
| 351 |
+
"learning_rate": 1.7191269385410685e-05,
|
| 352 |
+
"loss": 0.3913,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.2871912693854107,
|
| 357 |
+
"grad_norm": 8.942373275756836,
|
| 358 |
+
"learning_rate": 1.71338311315336e-05,
|
| 359 |
+
"loss": 0.4454,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.2929350947731189,
|
| 364 |
+
"grad_norm": 7.566854476928711,
|
| 365 |
+
"learning_rate": 1.707639287765652e-05,
|
| 366 |
+
"loss": 0.4704,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.2986789201608271,
|
| 371 |
+
"grad_norm": 5.819560527801514,
|
| 372 |
+
"learning_rate": 1.7018954623779438e-05,
|
| 373 |
+
"loss": 0.4245,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.3044227455485353,
|
| 378 |
+
"grad_norm": 7.269554138183594,
|
| 379 |
+
"learning_rate": 1.6961516369902355e-05,
|
| 380 |
+
"loss": 0.523,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.31016657093624356,
|
| 385 |
+
"grad_norm": 4.569669246673584,
|
| 386 |
+
"learning_rate": 1.690407811602527e-05,
|
| 387 |
+
"loss": 0.5053,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.31591039632395174,
|
| 392 |
+
"grad_norm": 6.8373332023620605,
|
| 393 |
+
"learning_rate": 1.684663986214819e-05,
|
| 394 |
+
"loss": 0.4592,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.32165422171165997,
|
| 399 |
+
"grad_norm": 7.736797332763672,
|
| 400 |
+
"learning_rate": 1.678920160827111e-05,
|
| 401 |
+
"loss": 0.4756,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.3273980470993682,
|
| 406 |
+
"grad_norm": 5.857370853424072,
|
| 407 |
+
"learning_rate": 1.6731763354394028e-05,
|
| 408 |
+
"loss": 0.4906,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.3331418724870764,
|
| 413 |
+
"grad_norm": 4.2734785079956055,
|
| 414 |
+
"learning_rate": 1.6674325100516945e-05,
|
| 415 |
+
"loss": 0.4623,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.3388856978747846,
|
| 420 |
+
"grad_norm": 7.859753131866455,
|
| 421 |
+
"learning_rate": 1.6616886846639865e-05,
|
| 422 |
+
"loss": 0.4568,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.3446295232624928,
|
| 427 |
+
"grad_norm": 4.936352252960205,
|
| 428 |
+
"learning_rate": 1.655944859276278e-05,
|
| 429 |
+
"loss": 0.3774,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.35037334865020103,
|
| 434 |
+
"grad_norm": 7.164617538452148,
|
| 435 |
+
"learning_rate": 1.6502010338885698e-05,
|
| 436 |
+
"loss": 0.5373,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.35611717403790927,
|
| 441 |
+
"grad_norm": 7.279328346252441,
|
| 442 |
+
"learning_rate": 1.6444572085008618e-05,
|
| 443 |
+
"loss": 0.4183,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.36186099942561745,
|
| 448 |
+
"grad_norm": 8.805562973022461,
|
| 449 |
+
"learning_rate": 1.6387133831131535e-05,
|
| 450 |
+
"loss": 0.4232,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.3676048248133257,
|
| 455 |
+
"grad_norm": 6.055676460266113,
|
| 456 |
+
"learning_rate": 1.632969557725445e-05,
|
| 457 |
+
"loss": 0.4316,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.3733486502010339,
|
| 462 |
+
"grad_norm": 5.835373401641846,
|
| 463 |
+
"learning_rate": 1.627225732337737e-05,
|
| 464 |
+
"loss": 0.4573,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.3790924755887421,
|
| 469 |
+
"grad_norm": 2.7914931774139404,
|
| 470 |
+
"learning_rate": 1.6214819069500288e-05,
|
| 471 |
+
"loss": 0.3691,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.38483630097645033,
|
| 476 |
+
"grad_norm": 5.780324459075928,
|
| 477 |
+
"learning_rate": 1.6157380815623208e-05,
|
| 478 |
+
"loss": 0.3761,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.3905801263641585,
|
| 483 |
+
"grad_norm": 5.7374348640441895,
|
| 484 |
+
"learning_rate": 1.6099942561746125e-05,
|
| 485 |
+
"loss": 0.4276,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.39632395175186674,
|
| 490 |
+
"grad_norm": 4.771572589874268,
|
| 491 |
+
"learning_rate": 1.604250430786904e-05,
|
| 492 |
+
"loss": 0.4092,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.402067777139575,
|
| 497 |
+
"grad_norm": 7.6168928146362305,
|
| 498 |
+
"learning_rate": 1.598506605399196e-05,
|
| 499 |
+
"loss": 0.4651,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.40781160252728316,
|
| 504 |
+
"grad_norm": 6.272294998168945,
|
| 505 |
+
"learning_rate": 1.5927627800114878e-05,
|
| 506 |
+
"loss": 0.4107,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.4135554279149914,
|
| 511 |
+
"grad_norm": 8.060046195983887,
|
| 512 |
+
"learning_rate": 1.5870189546237794e-05,
|
| 513 |
+
"loss": 0.4117,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.41929925330269957,
|
| 518 |
+
"grad_norm": 7.290645122528076,
|
| 519 |
+
"learning_rate": 1.5812751292360714e-05,
|
| 520 |
+
"loss": 0.5005,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.4250430786904078,
|
| 525 |
+
"grad_norm": 8.02639102935791,
|
| 526 |
+
"learning_rate": 1.575531303848363e-05,
|
| 527 |
+
"loss": 0.4561,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.43078690407811604,
|
| 532 |
+
"grad_norm": 8.455704689025879,
|
| 533 |
+
"learning_rate": 1.5697874784606548e-05,
|
| 534 |
+
"loss": 0.4592,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.4365307294658242,
|
| 539 |
+
"grad_norm": 2.9264323711395264,
|
| 540 |
+
"learning_rate": 1.5640436530729468e-05,
|
| 541 |
+
"loss": 0.4532,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.44227455485353245,
|
| 546 |
+
"grad_norm": 5.246438980102539,
|
| 547 |
+
"learning_rate": 1.5582998276852384e-05,
|
| 548 |
+
"loss": 0.4931,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.4480183802412407,
|
| 553 |
+
"grad_norm": 3.869628667831421,
|
| 554 |
+
"learning_rate": 1.5525560022975304e-05,
|
| 555 |
+
"loss": 0.3476,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.45376220562894887,
|
| 560 |
+
"grad_norm": 4.742978572845459,
|
| 561 |
+
"learning_rate": 1.546812176909822e-05,
|
| 562 |
+
"loss": 0.4274,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.4595060310166571,
|
| 567 |
+
"grad_norm": 6.9326300621032715,
|
| 568 |
+
"learning_rate": 1.5410683515221138e-05,
|
| 569 |
+
"loss": 0.5299,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.4652498564043653,
|
| 574 |
+
"grad_norm": 6.16196346282959,
|
| 575 |
+
"learning_rate": 1.5353245261344058e-05,
|
| 576 |
+
"loss": 0.4805,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.4709936817920735,
|
| 581 |
+
"grad_norm": 3.637753486633301,
|
| 582 |
+
"learning_rate": 1.5295807007466974e-05,
|
| 583 |
+
"loss": 0.4393,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.47673750717978175,
|
| 588 |
+
"grad_norm": 3.867966413497925,
|
| 589 |
+
"learning_rate": 1.523836875358989e-05,
|
| 590 |
+
"loss": 0.4358,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.48248133256748993,
|
| 595 |
+
"grad_norm": 4.753304481506348,
|
| 596 |
+
"learning_rate": 1.5180930499712809e-05,
|
| 597 |
+
"loss": 0.4737,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.48822515795519816,
|
| 602 |
+
"grad_norm": 7.7434515953063965,
|
| 603 |
+
"learning_rate": 1.5123492245835727e-05,
|
| 604 |
+
"loss": 0.4466,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.4939689833429064,
|
| 609 |
+
"grad_norm": 6.001241683959961,
|
| 610 |
+
"learning_rate": 1.5066053991958644e-05,
|
| 611 |
+
"loss": 0.4016,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.4997128087306146,
|
| 616 |
+
"grad_norm": 7.710599422454834,
|
| 617 |
+
"learning_rate": 1.5008615738081564e-05,
|
| 618 |
+
"loss": 0.4758,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.5054566341183228,
|
| 623 |
+
"grad_norm": 5.139091491699219,
|
| 624 |
+
"learning_rate": 1.4951177484204482e-05,
|
| 625 |
+
"loss": 0.4019,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.511200459506031,
|
| 630 |
+
"grad_norm": 9.428766250610352,
|
| 631 |
+
"learning_rate": 1.4893739230327399e-05,
|
| 632 |
+
"loss": 0.3742,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.5169442848937392,
|
| 637 |
+
"grad_norm": 9.071430206298828,
|
| 638 |
+
"learning_rate": 1.4836300976450317e-05,
|
| 639 |
+
"loss": 0.4183,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.5226881102814475,
|
| 644 |
+
"grad_norm": 10.34457015991211,
|
| 645 |
+
"learning_rate": 1.4778862722573236e-05,
|
| 646 |
+
"loss": 0.437,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.5284319356691557,
|
| 651 |
+
"grad_norm": 7.01099967956543,
|
| 652 |
+
"learning_rate": 1.4721424468696152e-05,
|
| 653 |
+
"loss": 0.4421,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.5341757610568638,
|
| 658 |
+
"grad_norm": 3.454484701156616,
|
| 659 |
+
"learning_rate": 1.466398621481907e-05,
|
| 660 |
+
"loss": 0.4283,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.539919586444572,
|
| 665 |
+
"grad_norm": 5.86787223815918,
|
| 666 |
+
"learning_rate": 1.4606547960941987e-05,
|
| 667 |
+
"loss": 0.4094,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.5456634118322803,
|
| 672 |
+
"grad_norm": 8.41640567779541,
|
| 673 |
+
"learning_rate": 1.4549109707064906e-05,
|
| 674 |
+
"loss": 0.414,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.5514072372199885,
|
| 679 |
+
"grad_norm": 6.936724662780762,
|
| 680 |
+
"learning_rate": 1.4491671453187824e-05,
|
| 681 |
+
"loss": 0.3936,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.5571510626076968,
|
| 686 |
+
"grad_norm": 7.984805583953857,
|
| 687 |
+
"learning_rate": 1.4434233199310744e-05,
|
| 688 |
+
"loss": 0.4071,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.562894887995405,
|
| 693 |
+
"grad_norm": 4.381742477416992,
|
| 694 |
+
"learning_rate": 1.437679494543366e-05,
|
| 695 |
+
"loss": 0.3393,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.5686387133831131,
|
| 700 |
+
"grad_norm": 7.115252494812012,
|
| 701 |
+
"learning_rate": 1.4319356691556579e-05,
|
| 702 |
+
"loss": 0.5616,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.5743825387708213,
|
| 707 |
+
"grad_norm": 6.60570764541626,
|
| 708 |
+
"learning_rate": 1.4261918437679495e-05,
|
| 709 |
+
"loss": 0.4633,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.5801263641585296,
|
| 714 |
+
"grad_norm": 9.552061080932617,
|
| 715 |
+
"learning_rate": 1.4204480183802414e-05,
|
| 716 |
+
"loss": 0.4973,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.5858701895462378,
|
| 721 |
+
"grad_norm": 3.6467580795288086,
|
| 722 |
+
"learning_rate": 1.4147041929925332e-05,
|
| 723 |
+
"loss": 0.3958,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.591614014933946,
|
| 728 |
+
"grad_norm": 2.663799524307251,
|
| 729 |
+
"learning_rate": 1.4089603676048249e-05,
|
| 730 |
+
"loss": 0.3402,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.5973578403216542,
|
| 735 |
+
"grad_norm": 7.468900203704834,
|
| 736 |
+
"learning_rate": 1.4032165422171167e-05,
|
| 737 |
+
"loss": 0.3341,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.6031016657093624,
|
| 742 |
+
"grad_norm": 10.396268844604492,
|
| 743 |
+
"learning_rate": 1.3974727168294084e-05,
|
| 744 |
+
"loss": 0.5122,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.6088454910970706,
|
| 749 |
+
"grad_norm": 7.79450798034668,
|
| 750 |
+
"learning_rate": 1.3917288914417002e-05,
|
| 751 |
+
"loss": 0.5581,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.6145893164847789,
|
| 756 |
+
"grad_norm": 8.077397346496582,
|
| 757 |
+
"learning_rate": 1.385985066053992e-05,
|
| 758 |
+
"loss": 0.4247,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.6203331418724871,
|
| 763 |
+
"grad_norm": 8.327542304992676,
|
| 764 |
+
"learning_rate": 1.380241240666284e-05,
|
| 765 |
+
"loss": 0.4122,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.6260769672601952,
|
| 770 |
+
"grad_norm": 7.940774917602539,
|
| 771 |
+
"learning_rate": 1.3744974152785757e-05,
|
| 772 |
+
"loss": 0.5199,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.6318207926479035,
|
| 777 |
+
"grad_norm": 5.148271560668945,
|
| 778 |
+
"learning_rate": 1.3687535898908675e-05,
|
| 779 |
+
"loss": 0.4534,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.6375646180356117,
|
| 784 |
+
"grad_norm": 7.042996883392334,
|
| 785 |
+
"learning_rate": 1.3630097645031592e-05,
|
| 786 |
+
"loss": 0.4737,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.6433084434233199,
|
| 791 |
+
"grad_norm": 5.131284236907959,
|
| 792 |
+
"learning_rate": 1.357265939115451e-05,
|
| 793 |
+
"loss": 0.3637,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.6490522688110282,
|
| 798 |
+
"grad_norm": 10.73865795135498,
|
| 799 |
+
"learning_rate": 1.3515221137277428e-05,
|
| 800 |
+
"loss": 0.4152,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.6547960941987364,
|
| 805 |
+
"grad_norm": 6.061016082763672,
|
| 806 |
+
"learning_rate": 1.3457782883400345e-05,
|
| 807 |
+
"loss": 0.343,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.6605399195864445,
|
| 812 |
+
"grad_norm": 11.200201988220215,
|
| 813 |
+
"learning_rate": 1.3400344629523263e-05,
|
| 814 |
+
"loss": 0.4781,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.6662837449741528,
|
| 819 |
+
"grad_norm": 6.987539768218994,
|
| 820 |
+
"learning_rate": 1.334290637564618e-05,
|
| 821 |
+
"loss": 0.4046,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.672027570361861,
|
| 826 |
+
"grad_norm": 7.3787713050842285,
|
| 827 |
+
"learning_rate": 1.3285468121769098e-05,
|
| 828 |
+
"loss": 0.4136,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.6777713957495692,
|
| 833 |
+
"grad_norm": 8.829427719116211,
|
| 834 |
+
"learning_rate": 1.3228029867892018e-05,
|
| 835 |
+
"loss": 0.3807,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.6835152211372775,
|
| 840 |
+
"grad_norm": 9.648842811584473,
|
| 841 |
+
"learning_rate": 1.3170591614014937e-05,
|
| 842 |
+
"loss": 0.3273,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.6892590465249856,
|
| 847 |
+
"grad_norm": 7.307587146759033,
|
| 848 |
+
"learning_rate": 1.3113153360137853e-05,
|
| 849 |
+
"loss": 0.3351,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.6950028719126938,
|
| 854 |
+
"grad_norm": 6.445584297180176,
|
| 855 |
+
"learning_rate": 1.3055715106260772e-05,
|
| 856 |
+
"loss": 0.4776,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.7007466973004021,
|
| 861 |
+
"grad_norm": 7.078349590301514,
|
| 862 |
+
"learning_rate": 1.2998276852383688e-05,
|
| 863 |
+
"loss": 0.4438,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.7064905226881103,
|
| 868 |
+
"grad_norm": 9.63571834564209,
|
| 869 |
+
"learning_rate": 1.2940838598506606e-05,
|
| 870 |
+
"loss": 0.4027,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.7122343480758185,
|
| 875 |
+
"grad_norm": 3.2861080169677734,
|
| 876 |
+
"learning_rate": 1.2883400344629525e-05,
|
| 877 |
+
"loss": 0.3334,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.7179781734635267,
|
| 882 |
+
"grad_norm": 7.433917999267578,
|
| 883 |
+
"learning_rate": 1.2825962090752441e-05,
|
| 884 |
+
"loss": 0.4418,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.7237219988512349,
|
| 889 |
+
"grad_norm": 7.511765003204346,
|
| 890 |
+
"learning_rate": 1.276852383687536e-05,
|
| 891 |
+
"loss": 0.3693,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.7294658242389431,
|
| 896 |
+
"grad_norm": 9.38160228729248,
|
| 897 |
+
"learning_rate": 1.2711085582998276e-05,
|
| 898 |
+
"loss": 0.3989,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.7352096496266514,
|
| 903 |
+
"grad_norm": 2.012756586074829,
|
| 904 |
+
"learning_rate": 1.2653647329121195e-05,
|
| 905 |
+
"loss": 0.3867,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.7409534750143596,
|
| 910 |
+
"grad_norm": 6.777096748352051,
|
| 911 |
+
"learning_rate": 1.2596209075244115e-05,
|
| 912 |
+
"loss": 0.5528,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.7466973004020678,
|
| 917 |
+
"grad_norm": 6.928145885467529,
|
| 918 |
+
"learning_rate": 1.2538770821367033e-05,
|
| 919 |
+
"loss": 0.3403,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.752441125789776,
|
| 924 |
+
"grad_norm": 7.99967622756958,
|
| 925 |
+
"learning_rate": 1.248133256748995e-05,
|
| 926 |
+
"loss": 0.5006,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.7581849511774842,
|
| 931 |
+
"grad_norm": 4.8364033699035645,
|
| 932 |
+
"learning_rate": 1.2423894313612868e-05,
|
| 933 |
+
"loss": 0.4266,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.7639287765651924,
|
| 938 |
+
"grad_norm": 4.017831802368164,
|
| 939 |
+
"learning_rate": 1.2366456059735785e-05,
|
| 940 |
+
"loss": 0.3444,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.7696726019529007,
|
| 945 |
+
"grad_norm": 5.27449893951416,
|
| 946 |
+
"learning_rate": 1.2309017805858703e-05,
|
| 947 |
+
"loss": 0.3962,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.7754164273406089,
|
| 952 |
+
"grad_norm": 7.989853858947754,
|
| 953 |
+
"learning_rate": 1.2251579551981621e-05,
|
| 954 |
+
"loss": 0.4172,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.781160252728317,
|
| 959 |
+
"grad_norm": 5.878440856933594,
|
| 960 |
+
"learning_rate": 1.2194141298104538e-05,
|
| 961 |
+
"loss": 0.4723,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.7869040781160253,
|
| 966 |
+
"grad_norm": 9.140411376953125,
|
| 967 |
+
"learning_rate": 1.2136703044227456e-05,
|
| 968 |
+
"loss": 0.4012,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.7926479035037335,
|
| 973 |
+
"grad_norm": 3.9119012355804443,
|
| 974 |
+
"learning_rate": 1.2079264790350373e-05,
|
| 975 |
+
"loss": 0.3545,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.7983917288914417,
|
| 980 |
+
"grad_norm": 6.248933792114258,
|
| 981 |
+
"learning_rate": 1.2021826536473291e-05,
|
| 982 |
+
"loss": 0.4787,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.80413555427915,
|
| 987 |
+
"grad_norm": 8.478959083557129,
|
| 988 |
+
"learning_rate": 1.1964388282596211e-05,
|
| 989 |
+
"loss": 0.4077,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.8098793796668581,
|
| 994 |
+
"grad_norm": 6.234384059906006,
|
| 995 |
+
"learning_rate": 1.190695002871913e-05,
|
| 996 |
+
"loss": 0.4044,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.8156232050545663,
|
| 1001 |
+
"grad_norm": 5.093031883239746,
|
| 1002 |
+
"learning_rate": 1.1849511774842046e-05,
|
| 1003 |
+
"loss": 0.3298,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.8213670304422745,
|
| 1008 |
+
"grad_norm": 5.755350112915039,
|
| 1009 |
+
"learning_rate": 1.1792073520964964e-05,
|
| 1010 |
+
"loss": 0.4099,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.8271108558299828,
|
| 1015 |
+
"grad_norm": 9.269704818725586,
|
| 1016 |
+
"learning_rate": 1.1734635267087881e-05,
|
| 1017 |
+
"loss": 0.366,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.832854681217691,
|
| 1022 |
+
"grad_norm": 4.977533340454102,
|
| 1023 |
+
"learning_rate": 1.16771970132108e-05,
|
| 1024 |
+
"loss": 0.3156,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.8385985066053991,
|
| 1029 |
+
"grad_norm": 6.767063140869141,
|
| 1030 |
+
"learning_rate": 1.1619758759333718e-05,
|
| 1031 |
+
"loss": 0.3966,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.8443423319931074,
|
| 1036 |
+
"grad_norm": 6.855627536773682,
|
| 1037 |
+
"learning_rate": 1.1562320505456634e-05,
|
| 1038 |
+
"loss": 0.3488,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.8500861573808156,
|
| 1043 |
+
"grad_norm": 3.408679723739624,
|
| 1044 |
+
"learning_rate": 1.1504882251579552e-05,
|
| 1045 |
+
"loss": 0.4016,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.8558299827685238,
|
| 1050 |
+
"grad_norm": 8.43376636505127,
|
| 1051 |
+
"learning_rate": 1.1447443997702469e-05,
|
| 1052 |
+
"loss": 0.3951,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.8615738081562321,
|
| 1057 |
+
"grad_norm": 7.106573104858398,
|
| 1058 |
+
"learning_rate": 1.1390005743825389e-05,
|
| 1059 |
+
"loss": 0.3056,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.8673176335439403,
|
| 1064 |
+
"grad_norm": 3.373734474182129,
|
| 1065 |
+
"learning_rate": 1.1332567489948307e-05,
|
| 1066 |
+
"loss": 0.4548,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.8730614589316484,
|
| 1071 |
+
"grad_norm": 4.657841205596924,
|
| 1072 |
+
"learning_rate": 1.1275129236071226e-05,
|
| 1073 |
+
"loss": 0.36,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.8788052843193567,
|
| 1078 |
+
"grad_norm": 8.218329429626465,
|
| 1079 |
+
"learning_rate": 1.1217690982194142e-05,
|
| 1080 |
+
"loss": 0.4297,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.8845491097070649,
|
| 1085 |
+
"grad_norm": 7.709052562713623,
|
| 1086 |
+
"learning_rate": 1.116025272831706e-05,
|
| 1087 |
+
"loss": 0.3766,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.8902929350947731,
|
| 1092 |
+
"grad_norm": 6.875143527984619,
|
| 1093 |
+
"learning_rate": 1.1102814474439977e-05,
|
| 1094 |
+
"loss": 0.4106,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.8960367604824814,
|
| 1099 |
+
"grad_norm": 5.460892200469971,
|
| 1100 |
+
"learning_rate": 1.1045376220562896e-05,
|
| 1101 |
+
"loss": 0.3016,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.9017805858701895,
|
| 1106 |
+
"grad_norm": 3.4830429553985596,
|
| 1107 |
+
"learning_rate": 1.0987937966685814e-05,
|
| 1108 |
+
"loss": 0.371,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.9075244112578977,
|
| 1113 |
+
"grad_norm": 8.233579635620117,
|
| 1114 |
+
"learning_rate": 1.093049971280873e-05,
|
| 1115 |
+
"loss": 0.4591,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.913268236645606,
|
| 1120 |
+
"grad_norm": 7.001081466674805,
|
| 1121 |
+
"learning_rate": 1.0873061458931649e-05,
|
| 1122 |
+
"loss": 0.3856,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.9190120620333142,
|
| 1127 |
+
"grad_norm": 7.473963260650635,
|
| 1128 |
+
"learning_rate": 1.0815623205054565e-05,
|
| 1129 |
+
"loss": 0.4937,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.9247558874210224,
|
| 1134 |
+
"grad_norm": 4.046863079071045,
|
| 1135 |
+
"learning_rate": 1.0758184951177485e-05,
|
| 1136 |
+
"loss": 0.4141,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.9304997128087306,
|
| 1141 |
+
"grad_norm": 5.425885200500488,
|
| 1142 |
+
"learning_rate": 1.0700746697300404e-05,
|
| 1143 |
+
"loss": 0.3545,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.9362435381964388,
|
| 1148 |
+
"grad_norm": 5.255967140197754,
|
| 1149 |
+
"learning_rate": 1.0643308443423322e-05,
|
| 1150 |
+
"loss": 0.3881,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.941987363584147,
|
| 1155 |
+
"grad_norm": 7.365703105926514,
|
| 1156 |
+
"learning_rate": 1.0585870189546239e-05,
|
| 1157 |
+
"loss": 0.3648,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.9477311889718553,
|
| 1162 |
+
"grad_norm": 5.149658679962158,
|
| 1163 |
+
"learning_rate": 1.0528431935669157e-05,
|
| 1164 |
+
"loss": 0.2952,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.9534750143595635,
|
| 1169 |
+
"grad_norm": 5.5968194007873535,
|
| 1170 |
+
"learning_rate": 1.0470993681792074e-05,
|
| 1171 |
+
"loss": 0.3604,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.9592188397472717,
|
| 1176 |
+
"grad_norm": 6.176368713378906,
|
| 1177 |
+
"learning_rate": 1.0413555427914992e-05,
|
| 1178 |
+
"loss": 0.3584,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.9649626651349799,
|
| 1183 |
+
"grad_norm": 5.876338958740234,
|
| 1184 |
+
"learning_rate": 1.035611717403791e-05,
|
| 1185 |
+
"loss": 0.4076,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.9707064905226881,
|
| 1190 |
+
"grad_norm": 11.697908401489258,
|
| 1191 |
+
"learning_rate": 1.0298678920160827e-05,
|
| 1192 |
+
"loss": 0.3648,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.9764503159103963,
|
| 1197 |
+
"grad_norm": 7.04163122177124,
|
| 1198 |
+
"learning_rate": 1.0241240666283745e-05,
|
| 1199 |
+
"loss": 0.349,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.9821941412981046,
|
| 1204 |
+
"grad_norm": 6.7707133293151855,
|
| 1205 |
+
"learning_rate": 1.0183802412406662e-05,
|
| 1206 |
+
"loss": 0.3888,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.9879379666858128,
|
| 1211 |
+
"grad_norm": 3.8108270168304443,
|
| 1212 |
+
"learning_rate": 1.0126364158529582e-05,
|
| 1213 |
+
"loss": 0.3753,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.9936817920735209,
|
| 1218 |
+
"grad_norm": 11.013320922851562,
|
| 1219 |
+
"learning_rate": 1.00689259046525e-05,
|
| 1220 |
+
"loss": 0.3977,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.9994256174612292,
|
| 1225 |
+
"grad_norm": 4.042791843414307,
|
| 1226 |
+
"learning_rate": 1.0011487650775419e-05,
|
| 1227 |
+
"loss": 0.3724,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.0051694428489375,
|
| 1232 |
+
"grad_norm": 6.258309841156006,
|
| 1233 |
+
"learning_rate": 9.954049396898335e-06,
|
| 1234 |
+
"loss": 0.3591,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.0109132682366455,
|
| 1239 |
+
"grad_norm": 7.884782314300537,
|
| 1240 |
+
"learning_rate": 9.896611143021253e-06,
|
| 1241 |
+
"loss": 0.4114,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.0166570936243537,
|
| 1246 |
+
"grad_norm": 4.9663567543029785,
|
| 1247 |
+
"learning_rate": 9.83917288914417e-06,
|
| 1248 |
+
"loss": 0.4462,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.022400919012062,
|
| 1253 |
+
"grad_norm": 7.046320915222168,
|
| 1254 |
+
"learning_rate": 9.781734635267088e-06,
|
| 1255 |
+
"loss": 0.3386,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.0281447443997702,
|
| 1260 |
+
"grad_norm": 6.846945762634277,
|
| 1261 |
+
"learning_rate": 9.724296381390007e-06,
|
| 1262 |
+
"loss": 0.3181,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.0338885697874785,
|
| 1267 |
+
"grad_norm": 5.925526142120361,
|
| 1268 |
+
"learning_rate": 9.666858127512925e-06,
|
| 1269 |
+
"loss": 0.3357,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.0396323951751867,
|
| 1274 |
+
"grad_norm": 14.302725791931152,
|
| 1275 |
+
"learning_rate": 9.609419873635842e-06,
|
| 1276 |
+
"loss": 0.2859,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.045376220562895,
|
| 1281 |
+
"grad_norm": 8.27291488647461,
|
| 1282 |
+
"learning_rate": 9.55198161975876e-06,
|
| 1283 |
+
"loss": 0.3688,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.0511200459506032,
|
| 1288 |
+
"grad_norm": 5.266950607299805,
|
| 1289 |
+
"learning_rate": 9.494543365881678e-06,
|
| 1290 |
+
"loss": 0.3106,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.0568638713383114,
|
| 1295 |
+
"grad_norm": 11.347005844116211,
|
| 1296 |
+
"learning_rate": 9.437105112004595e-06,
|
| 1297 |
+
"loss": 0.3389,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.0626076967260196,
|
| 1302 |
+
"grad_norm": 4.7072906494140625,
|
| 1303 |
+
"learning_rate": 9.379666858127515e-06,
|
| 1304 |
+
"loss": 0.3519,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.0683515221137276,
|
| 1309 |
+
"grad_norm": 4.05309534072876,
|
| 1310 |
+
"learning_rate": 9.322228604250432e-06,
|
| 1311 |
+
"loss": 0.3006,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.0740953475014359,
|
| 1316 |
+
"grad_norm": 5.578520774841309,
|
| 1317 |
+
"learning_rate": 9.26479035037335e-06,
|
| 1318 |
+
"loss": 0.3645,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.079839172889144,
|
| 1323 |
+
"grad_norm": 7.405791282653809,
|
| 1324 |
+
"learning_rate": 9.207352096496266e-06,
|
| 1325 |
+
"loss": 0.3104,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.0855829982768523,
|
| 1330 |
+
"grad_norm": 9.269173622131348,
|
| 1331 |
+
"learning_rate": 9.149913842619185e-06,
|
| 1332 |
+
"loss": 0.3576,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.0913268236645606,
|
| 1337 |
+
"grad_norm": 5.276297569274902,
|
| 1338 |
+
"learning_rate": 9.092475588742103e-06,
|
| 1339 |
+
"loss": 0.3354,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.0970706490522688,
|
| 1344 |
+
"grad_norm": 8.320406913757324,
|
| 1345 |
+
"learning_rate": 9.035037334865021e-06,
|
| 1346 |
+
"loss": 0.3232,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.102814474439977,
|
| 1351 |
+
"grad_norm": 6.023215293884277,
|
| 1352 |
+
"learning_rate": 8.977599080987938e-06,
|
| 1353 |
+
"loss": 0.3427,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.1085582998276853,
|
| 1358 |
+
"grad_norm": 8.178590774536133,
|
| 1359 |
+
"learning_rate": 8.920160827110856e-06,
|
| 1360 |
+
"loss": 0.3103,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.1143021252153935,
|
| 1365 |
+
"grad_norm": 6.056619644165039,
|
| 1366 |
+
"learning_rate": 8.862722573233775e-06,
|
| 1367 |
+
"loss": 0.3848,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.1200459506031017,
|
| 1372 |
+
"grad_norm": 6.109485626220703,
|
| 1373 |
+
"learning_rate": 8.805284319356693e-06,
|
| 1374 |
+
"loss": 0.3102,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.12578977599081,
|
| 1379 |
+
"grad_norm": 6.949984550476074,
|
| 1380 |
+
"learning_rate": 8.747846065479611e-06,
|
| 1381 |
+
"loss": 0.3013,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.1315336013785182,
|
| 1386 |
+
"grad_norm": 4.320880889892578,
|
| 1387 |
+
"learning_rate": 8.690407811602528e-06,
|
| 1388 |
+
"loss": 0.3169,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.1372774267662262,
|
| 1393 |
+
"grad_norm": 9.62964916229248,
|
| 1394 |
+
"learning_rate": 8.632969557725446e-06,
|
| 1395 |
+
"loss": 0.3577,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.1430212521539345,
|
| 1400 |
+
"grad_norm": 7.1865105628967285,
|
| 1401 |
+
"learning_rate": 8.575531303848363e-06,
|
| 1402 |
+
"loss": 0.4245,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.1487650775416427,
|
| 1407 |
+
"grad_norm": 11.42944622039795,
|
| 1408 |
+
"learning_rate": 8.518093049971281e-06,
|
| 1409 |
+
"loss": 0.3421,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.154508902929351,
|
| 1414 |
+
"grad_norm": 10.365814208984375,
|
| 1415 |
+
"learning_rate": 8.4606547960942e-06,
|
| 1416 |
+
"loss": 0.2971,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.1602527283170592,
|
| 1421 |
+
"grad_norm": 4.546888828277588,
|
| 1422 |
+
"learning_rate": 8.403216542217118e-06,
|
| 1423 |
+
"loss": 0.3762,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.1659965537047674,
|
| 1428 |
+
"grad_norm": 9.672823905944824,
|
| 1429 |
+
"learning_rate": 8.345778288340034e-06,
|
| 1430 |
+
"loss": 0.3411,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.1717403790924756,
|
| 1435 |
+
"grad_norm": 4.738915920257568,
|
| 1436 |
+
"learning_rate": 8.288340034462953e-06,
|
| 1437 |
+
"loss": 0.3453,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.1774842044801839,
|
| 1442 |
+
"grad_norm": 10.187810897827148,
|
| 1443 |
+
"learning_rate": 8.230901780585871e-06,
|
| 1444 |
+
"loss": 0.3351,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 1.183228029867892,
|
| 1449 |
+
"grad_norm": 6.290671348571777,
|
| 1450 |
+
"learning_rate": 8.17346352670879e-06,
|
| 1451 |
+
"loss": 0.3286,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.1889718552556001,
|
| 1456 |
+
"grad_norm": 9.14261531829834,
|
| 1457 |
+
"learning_rate": 8.116025272831708e-06,
|
| 1458 |
+
"loss": 0.2878,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.1947156806433084,
|
| 1463 |
+
"grad_norm": 7.814758777618408,
|
| 1464 |
+
"learning_rate": 8.058587018954624e-06,
|
| 1465 |
+
"loss": 0.3469,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.2004595060310166,
|
| 1470 |
+
"grad_norm": 10.085731506347656,
|
| 1471 |
+
"learning_rate": 8.001148765077543e-06,
|
| 1472 |
+
"loss": 0.3025,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.2062033314187248,
|
| 1477 |
+
"grad_norm": 10.734376907348633,
|
| 1478 |
+
"learning_rate": 7.94371051120046e-06,
|
| 1479 |
+
"loss": 0.2442,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.211947156806433,
|
| 1484 |
+
"grad_norm": 13.61286735534668,
|
| 1485 |
+
"learning_rate": 7.88627225732338e-06,
|
| 1486 |
+
"loss": 0.3063,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.2176909821941413,
|
| 1491 |
+
"grad_norm": 8.572850227355957,
|
| 1492 |
+
"learning_rate": 7.828834003446296e-06,
|
| 1493 |
+
"loss": 0.3863,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 1.2234348075818495,
|
| 1498 |
+
"grad_norm": 6.247170448303223,
|
| 1499 |
+
"learning_rate": 7.771395749569214e-06,
|
| 1500 |
+
"loss": 0.3214,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 1.2291786329695578,
|
| 1505 |
+
"grad_norm": 7.438636779785156,
|
| 1506 |
+
"learning_rate": 7.71395749569213e-06,
|
| 1507 |
+
"loss": 0.381,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 1.234922458357266,
|
| 1512 |
+
"grad_norm": 6.11846399307251,
|
| 1513 |
+
"learning_rate": 7.656519241815049e-06,
|
| 1514 |
+
"loss": 0.3556,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 1.2406662837449742,
|
| 1519 |
+
"grad_norm": 10.697092056274414,
|
| 1520 |
+
"learning_rate": 7.5990809879379666e-06,
|
| 1521 |
+
"loss": 0.3767,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 1.2464101091326825,
|
| 1526 |
+
"grad_norm": 5.3118205070495605,
|
| 1527 |
+
"learning_rate": 7.541642734060886e-06,
|
| 1528 |
+
"loss": 0.3112,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 1.2521539345203907,
|
| 1533 |
+
"grad_norm": 5.907925128936768,
|
| 1534 |
+
"learning_rate": 7.484204480183803e-06,
|
| 1535 |
+
"loss": 0.2833,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 1.2578977599080987,
|
| 1540 |
+
"grad_norm": 7.271302223205566,
|
| 1541 |
+
"learning_rate": 7.426766226306721e-06,
|
| 1542 |
+
"loss": 0.1935,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 1.263641585295807,
|
| 1547 |
+
"grad_norm": 12.389423370361328,
|
| 1548 |
+
"learning_rate": 7.369327972429638e-06,
|
| 1549 |
+
"loss": 0.3946,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.2693854106835152,
|
| 1554 |
+
"grad_norm": 9.09422492980957,
|
| 1555 |
+
"learning_rate": 7.3118897185525564e-06,
|
| 1556 |
+
"loss": 0.3191,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.2751292360712234,
|
| 1561 |
+
"grad_norm": 8.75156307220459,
|
| 1562 |
+
"learning_rate": 7.254451464675475e-06,
|
| 1563 |
+
"loss": 0.4077,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.2808730614589316,
|
| 1568 |
+
"grad_norm": 7.306863784790039,
|
| 1569 |
+
"learning_rate": 7.197013210798392e-06,
|
| 1570 |
+
"loss": 0.3295,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.2866168868466399,
|
| 1575 |
+
"grad_norm": 9.715473175048828,
|
| 1576 |
+
"learning_rate": 7.1395749569213105e-06,
|
| 1577 |
+
"loss": 0.4163,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 1.2923607122343481,
|
| 1582 |
+
"grad_norm": 6.315252780914307,
|
| 1583 |
+
"learning_rate": 7.082136703044228e-06,
|
| 1584 |
+
"loss": 0.3817,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.2981045376220564,
|
| 1589 |
+
"grad_norm": 8.821859359741211,
|
| 1590 |
+
"learning_rate": 7.0246984491671455e-06,
|
| 1591 |
+
"loss": 0.3265,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.3038483630097644,
|
| 1596 |
+
"grad_norm": 6.838233947753906,
|
| 1597 |
+
"learning_rate": 6.967260195290065e-06,
|
| 1598 |
+
"loss": 0.2966,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 1.3095921883974726,
|
| 1603 |
+
"grad_norm": 9.925073623657227,
|
| 1604 |
+
"learning_rate": 6.909821941412982e-06,
|
| 1605 |
+
"loss": 0.4484,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.3153360137851808,
|
| 1610 |
+
"grad_norm": 5.026411056518555,
|
| 1611 |
+
"learning_rate": 6.8523836875358996e-06,
|
| 1612 |
+
"loss": 0.4647,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 1.321079839172889,
|
| 1617 |
+
"grad_norm": 4.3956732749938965,
|
| 1618 |
+
"learning_rate": 6.794945433658817e-06,
|
| 1619 |
+
"loss": 0.2968,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 1.3268236645605973,
|
| 1624 |
+
"grad_norm": 6.904971599578857,
|
| 1625 |
+
"learning_rate": 6.7375071797817345e-06,
|
| 1626 |
+
"loss": 0.2888,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 1.3325674899483055,
|
| 1631 |
+
"grad_norm": 3.1684281826019287,
|
| 1632 |
+
"learning_rate": 6.680068925904653e-06,
|
| 1633 |
+
"loss": 0.2975,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 1.3383113153360138,
|
| 1638 |
+
"grad_norm": 7.3333420753479,
|
| 1639 |
+
"learning_rate": 6.622630672027571e-06,
|
| 1640 |
+
"loss": 0.3911,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 1.344055140723722,
|
| 1645 |
+
"grad_norm": 7.822445392608643,
|
| 1646 |
+
"learning_rate": 6.565192418150489e-06,
|
| 1647 |
+
"loss": 0.3199,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 1.3497989661114302,
|
| 1652 |
+
"grad_norm": 9.02872371673584,
|
| 1653 |
+
"learning_rate": 6.507754164273407e-06,
|
| 1654 |
+
"loss": 0.4698,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 1.3555427914991385,
|
| 1659 |
+
"grad_norm": 3.9332520961761475,
|
| 1660 |
+
"learning_rate": 6.450315910396324e-06,
|
| 1661 |
+
"loss": 0.3651,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 1.3612866168868467,
|
| 1666 |
+
"grad_norm": 7.590347766876221,
|
| 1667 |
+
"learning_rate": 6.392877656519242e-06,
|
| 1668 |
+
"loss": 0.3121,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 1.367030442274555,
|
| 1673 |
+
"grad_norm": 8.964584350585938,
|
| 1674 |
+
"learning_rate": 6.335439402642161e-06,
|
| 1675 |
+
"loss": 0.271,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 1.3727742676622632,
|
| 1680 |
+
"grad_norm": 8.058918952941895,
|
| 1681 |
+
"learning_rate": 6.2780011487650785e-06,
|
| 1682 |
+
"loss": 0.3906,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 1.3785180930499714,
|
| 1687 |
+
"grad_norm": 6.742099761962891,
|
| 1688 |
+
"learning_rate": 6.220562894887996e-06,
|
| 1689 |
+
"loss": 0.3072,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 1.3842619184376794,
|
| 1694 |
+
"grad_norm": 5.961569309234619,
|
| 1695 |
+
"learning_rate": 6.1631246410109134e-06,
|
| 1696 |
+
"loss": 0.3673,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 1.3900057438253877,
|
| 1701 |
+
"grad_norm": 9.705893516540527,
|
| 1702 |
+
"learning_rate": 6.105686387133831e-06,
|
| 1703 |
+
"loss": 0.3544,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 1.395749569213096,
|
| 1708 |
+
"grad_norm": 4.435375690460205,
|
| 1709 |
+
"learning_rate": 6.04824813325675e-06,
|
| 1710 |
+
"loss": 0.2372,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 1.4014933946008041,
|
| 1715 |
+
"grad_norm": 5.375720977783203,
|
| 1716 |
+
"learning_rate": 5.9908098793796675e-06,
|
| 1717 |
+
"loss": 0.2264,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 1.4072372199885124,
|
| 1722 |
+
"grad_norm": 5.602358818054199,
|
| 1723 |
+
"learning_rate": 5.933371625502585e-06,
|
| 1724 |
+
"loss": 0.3449,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 1.4129810453762206,
|
| 1729 |
+
"grad_norm": 10.811373710632324,
|
| 1730 |
+
"learning_rate": 5.875933371625503e-06,
|
| 1731 |
+
"loss": 0.3663,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 1.4187248707639288,
|
| 1736 |
+
"grad_norm": 10.196518898010254,
|
| 1737 |
+
"learning_rate": 5.818495117748421e-06,
|
| 1738 |
+
"loss": 0.3546,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 1.424468696151637,
|
| 1743 |
+
"grad_norm": 10.06306266784668,
|
| 1744 |
+
"learning_rate": 5.761056863871339e-06,
|
| 1745 |
+
"loss": 0.3282,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 1.430212521539345,
|
| 1750 |
+
"grad_norm": 4.978325843811035,
|
| 1751 |
+
"learning_rate": 5.703618609994257e-06,
|
| 1752 |
+
"loss": 0.2961,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 1.4359563469270533,
|
| 1757 |
+
"grad_norm": 10.731146812438965,
|
| 1758 |
+
"learning_rate": 5.646180356117175e-06,
|
| 1759 |
+
"loss": 0.3349,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 1.4417001723147616,
|
| 1764 |
+
"grad_norm": 8.913891792297363,
|
| 1765 |
+
"learning_rate": 5.588742102240092e-06,
|
| 1766 |
+
"loss": 0.3131,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 1.4474439977024698,
|
| 1771 |
+
"grad_norm": 5.1745195388793945,
|
| 1772 |
+
"learning_rate": 5.53130384836301e-06,
|
| 1773 |
+
"loss": 0.3842,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 1.453187823090178,
|
| 1778 |
+
"grad_norm": 8.361491203308105,
|
| 1779 |
+
"learning_rate": 5.473865594485927e-06,
|
| 1780 |
+
"loss": 0.3598,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 1.4589316484778863,
|
| 1785 |
+
"grad_norm": 6.487078666687012,
|
| 1786 |
+
"learning_rate": 5.4164273406088464e-06,
|
| 1787 |
+
"loss": 0.3244,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 1.4646754738655945,
|
| 1792 |
+
"grad_norm": 4.129726409912109,
|
| 1793 |
+
"learning_rate": 5.358989086731764e-06,
|
| 1794 |
+
"loss": 0.3152,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 1.4704192992533027,
|
| 1799 |
+
"grad_norm": 9.363592147827148,
|
| 1800 |
+
"learning_rate": 5.301550832854681e-06,
|
| 1801 |
+
"loss": 0.3321,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 1.476163124641011,
|
| 1806 |
+
"grad_norm": 6.334773063659668,
|
| 1807 |
+
"learning_rate": 5.2441125789776e-06,
|
| 1808 |
+
"loss": 0.3312,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 1.4819069500287192,
|
| 1813 |
+
"grad_norm": 7.404930114746094,
|
| 1814 |
+
"learning_rate": 5.186674325100517e-06,
|
| 1815 |
+
"loss": 0.3739,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 1.4876507754164274,
|
| 1820 |
+
"grad_norm": 7.487016201019287,
|
| 1821 |
+
"learning_rate": 5.1292360712234355e-06,
|
| 1822 |
+
"loss": 0.2865,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 1.4933946008041357,
|
| 1827 |
+
"grad_norm": 13.322307586669922,
|
| 1828 |
+
"learning_rate": 5.071797817346353e-06,
|
| 1829 |
+
"loss": 0.3444,
|
| 1830 |
+
"step": 2600
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 1.499138426191844,
|
| 1834 |
+
"grad_norm": 9.053878784179688,
|
| 1835 |
+
"learning_rate": 5.014359563469271e-06,
|
| 1836 |
+
"loss": 0.3799,
|
| 1837 |
+
"step": 2610
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 1.5048822515795521,
|
| 1841 |
+
"grad_norm": 4.018943786621094,
|
| 1842 |
+
"learning_rate": 4.956921309592189e-06,
|
| 1843 |
+
"loss": 0.2567,
|
| 1844 |
+
"step": 2620
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 1.5106260769672601,
|
| 1848 |
+
"grad_norm": 2.2457354068756104,
|
| 1849 |
+
"learning_rate": 4.899483055715107e-06,
|
| 1850 |
+
"loss": 0.2978,
|
| 1851 |
+
"step": 2630
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 1.5163699023549684,
|
| 1855 |
+
"grad_norm": 4.894889831542969,
|
| 1856 |
+
"learning_rate": 4.8420448018380245e-06,
|
| 1857 |
+
"loss": 0.2861,
|
| 1858 |
+
"step": 2640
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 1.5221137277426766,
|
| 1862 |
+
"grad_norm": 6.843629360198975,
|
| 1863 |
+
"learning_rate": 4.784606547960942e-06,
|
| 1864 |
+
"loss": 0.3254,
|
| 1865 |
+
"step": 2650
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 1.5278575531303848,
|
| 1869 |
+
"grad_norm": 7.173573970794678,
|
| 1870 |
+
"learning_rate": 4.72716829408386e-06,
|
| 1871 |
+
"loss": 0.3481,
|
| 1872 |
+
"step": 2660
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 1.533601378518093,
|
| 1876 |
+
"grad_norm": 12.328543663024902,
|
| 1877 |
+
"learning_rate": 4.669730040206778e-06,
|
| 1878 |
+
"loss": 0.3124,
|
| 1879 |
+
"step": 2670
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 1.5393452039058013,
|
| 1883 |
+
"grad_norm": 9.337592124938965,
|
| 1884 |
+
"learning_rate": 4.612291786329696e-06,
|
| 1885 |
+
"loss": 0.3005,
|
| 1886 |
+
"step": 2680
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 1.5450890292935093,
|
| 1890 |
+
"grad_norm": 5.477969646453857,
|
| 1891 |
+
"learning_rate": 4.5548535324526135e-06,
|
| 1892 |
+
"loss": 0.3457,
|
| 1893 |
+
"step": 2690
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 1.5508328546812176,
|
| 1897 |
+
"grad_norm": 5.083920955657959,
|
| 1898 |
+
"learning_rate": 4.497415278575532e-06,
|
| 1899 |
+
"loss": 0.2858,
|
| 1900 |
+
"step": 2700
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 1.5565766800689258,
|
| 1904 |
+
"grad_norm": 6.250855445861816,
|
| 1905 |
+
"learning_rate": 4.439977024698449e-06,
|
| 1906 |
+
"loss": 0.2673,
|
| 1907 |
+
"step": 2710
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 1.562320505456634,
|
| 1911 |
+
"grad_norm": 6.169952392578125,
|
| 1912 |
+
"learning_rate": 4.382538770821368e-06,
|
| 1913 |
+
"loss": 0.2971,
|
| 1914 |
+
"step": 2720
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 1.5680643308443423,
|
| 1918 |
+
"grad_norm": 8.261754989624023,
|
| 1919 |
+
"learning_rate": 4.325100516944285e-06,
|
| 1920 |
+
"loss": 0.3009,
|
| 1921 |
+
"step": 2730
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 1.5738081562320505,
|
| 1925 |
+
"grad_norm": 8.477384567260742,
|
| 1926 |
+
"learning_rate": 4.267662263067203e-06,
|
| 1927 |
+
"loss": 0.3017,
|
| 1928 |
+
"step": 2740
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 1.5795519816197587,
|
| 1932 |
+
"grad_norm": 8.52374267578125,
|
| 1933 |
+
"learning_rate": 4.210224009190121e-06,
|
| 1934 |
+
"loss": 0.4511,
|
| 1935 |
+
"step": 2750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 1.585295807007467,
|
| 1939 |
+
"grad_norm": 8.646858215332031,
|
| 1940 |
+
"learning_rate": 4.152785755313039e-06,
|
| 1941 |
+
"loss": 0.2699,
|
| 1942 |
+
"step": 2760
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 1.5910396323951752,
|
| 1946 |
+
"grad_norm": 7.500174522399902,
|
| 1947 |
+
"learning_rate": 4.095347501435957e-06,
|
| 1948 |
+
"loss": 0.2743,
|
| 1949 |
+
"step": 2770
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 1.5967834577828834,
|
| 1953 |
+
"grad_norm": 5.454465389251709,
|
| 1954 |
+
"learning_rate": 4.037909247558874e-06,
|
| 1955 |
+
"loss": 0.3023,
|
| 1956 |
+
"step": 2780
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 1.6025272831705917,
|
| 1960 |
+
"grad_norm": 2.6998019218444824,
|
| 1961 |
+
"learning_rate": 3.9804709936817925e-06,
|
| 1962 |
+
"loss": 0.3941,
|
| 1963 |
+
"step": 2790
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 1.6082711085583,
|
| 1967 |
+
"grad_norm": 4.594570159912109,
|
| 1968 |
+
"learning_rate": 3.92303273980471e-06,
|
| 1969 |
+
"loss": 0.252,
|
| 1970 |
+
"step": 2800
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 1.6140149339460081,
|
| 1974 |
+
"grad_norm": 5.87538480758667,
|
| 1975 |
+
"learning_rate": 3.865594485927628e-06,
|
| 1976 |
+
"loss": 0.3093,
|
| 1977 |
+
"step": 2810
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 1.6197587593337164,
|
| 1981 |
+
"grad_norm": 5.358250617980957,
|
| 1982 |
+
"learning_rate": 3.808156232050546e-06,
|
| 1983 |
+
"loss": 0.3674,
|
| 1984 |
+
"step": 2820
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 1.6255025847214246,
|
| 1988 |
+
"grad_norm": 4.871222972869873,
|
| 1989 |
+
"learning_rate": 3.7507179781734636e-06,
|
| 1990 |
+
"loss": 0.2407,
|
| 1991 |
+
"step": 2830
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 1.6312464101091326,
|
| 1995 |
+
"grad_norm": 6.6836838722229,
|
| 1996 |
+
"learning_rate": 3.693279724296382e-06,
|
| 1997 |
+
"loss": 0.2978,
|
| 1998 |
+
"step": 2840
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 1.6369902354968409,
|
| 2002 |
+
"grad_norm": 7.67780065536499,
|
| 2003 |
+
"learning_rate": 3.6358414704192994e-06,
|
| 2004 |
+
"loss": 0.3655,
|
| 2005 |
+
"step": 2850
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 1.642734060884549,
|
| 2009 |
+
"grad_norm": 6.890137672424316,
|
| 2010 |
+
"learning_rate": 3.5784032165422173e-06,
|
| 2011 |
+
"loss": 0.3618,
|
| 2012 |
+
"step": 2860
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 1.6484778862722573,
|
| 2016 |
+
"grad_norm": 7.769665241241455,
|
| 2017 |
+
"learning_rate": 3.5209649626651356e-06,
|
| 2018 |
+
"loss": 0.2703,
|
| 2019 |
+
"step": 2870
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 1.6542217116599656,
|
| 2023 |
+
"grad_norm": 4.888299465179443,
|
| 2024 |
+
"learning_rate": 3.463526708788053e-06,
|
| 2025 |
+
"loss": 0.3257,
|
| 2026 |
+
"step": 2880
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 1.6599655370476738,
|
| 2030 |
+
"grad_norm": 4.726266384124756,
|
| 2031 |
+
"learning_rate": 3.406088454910971e-06,
|
| 2032 |
+
"loss": 0.3573,
|
| 2033 |
+
"step": 2890
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 1.6657093624353818,
|
| 2037 |
+
"grad_norm": 6.836297035217285,
|
| 2038 |
+
"learning_rate": 3.348650201033889e-06,
|
| 2039 |
+
"loss": 0.3325,
|
| 2040 |
+
"step": 2900
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 1.67145318782309,
|
| 2044 |
+
"grad_norm": 6.571508884429932,
|
| 2045 |
+
"learning_rate": 3.2912119471568067e-06,
|
| 2046 |
+
"loss": 0.292,
|
| 2047 |
+
"step": 2910
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 1.6771970132107983,
|
| 2051 |
+
"grad_norm": 9.843769073486328,
|
| 2052 |
+
"learning_rate": 3.2337736932797246e-06,
|
| 2053 |
+
"loss": 0.4027,
|
| 2054 |
+
"step": 2920
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 1.6829408385985065,
|
| 2058 |
+
"grad_norm": 6.089911937713623,
|
| 2059 |
+
"learning_rate": 3.1763354394026425e-06,
|
| 2060 |
+
"loss": 0.3627,
|
| 2061 |
+
"step": 2930
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 1.6886846639862148,
|
| 2065 |
+
"grad_norm": 6.846927165985107,
|
| 2066 |
+
"learning_rate": 3.11889718552556e-06,
|
| 2067 |
+
"loss": 0.3541,
|
| 2068 |
+
"step": 2940
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 1.694428489373923,
|
| 2072 |
+
"grad_norm": 10.23181438446045,
|
| 2073 |
+
"learning_rate": 3.0614589316484783e-06,
|
| 2074 |
+
"loss": 0.3418,
|
| 2075 |
+
"step": 2950
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 1.7001723147616312,
|
| 2079 |
+
"grad_norm": 5.403523921966553,
|
| 2080 |
+
"learning_rate": 3.0040206777713958e-06,
|
| 2081 |
+
"loss": 0.3498,
|
| 2082 |
+
"step": 2960
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 1.7059161401493395,
|
| 2086 |
+
"grad_norm": 8.252917289733887,
|
| 2087 |
+
"learning_rate": 2.9465824238943137e-06,
|
| 2088 |
+
"loss": 0.3401,
|
| 2089 |
+
"step": 2970
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 1.7116599655370477,
|
| 2093 |
+
"grad_norm": 7.06523323059082,
|
| 2094 |
+
"learning_rate": 2.889144170017232e-06,
|
| 2095 |
+
"loss": 0.2331,
|
| 2096 |
+
"step": 2980
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 1.717403790924756,
|
| 2100 |
+
"grad_norm": 7.739984035491943,
|
| 2101 |
+
"learning_rate": 2.8317059161401494e-06,
|
| 2102 |
+
"loss": 0.3678,
|
| 2103 |
+
"step": 2990
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 1.7231476163124642,
|
| 2107 |
+
"grad_norm": 4.021157741546631,
|
| 2108 |
+
"learning_rate": 2.7742676622630677e-06,
|
| 2109 |
+
"loss": 0.3537,
|
| 2110 |
+
"step": 3000
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 1.7288914417001724,
|
| 2114 |
+
"grad_norm": 9.555063247680664,
|
| 2115 |
+
"learning_rate": 2.7168294083859852e-06,
|
| 2116 |
+
"loss": 0.2955,
|
| 2117 |
+
"step": 3010
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 1.7346352670878806,
|
| 2121 |
+
"grad_norm": 6.779582500457764,
|
| 2122 |
+
"learning_rate": 2.659391154508903e-06,
|
| 2123 |
+
"loss": 0.3139,
|
| 2124 |
+
"step": 3020
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 1.7403790924755889,
|
| 2128 |
+
"grad_norm": 2.2541654109954834,
|
| 2129 |
+
"learning_rate": 2.601952900631821e-06,
|
| 2130 |
+
"loss": 0.2643,
|
| 2131 |
+
"step": 3030
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 1.746122917863297,
|
| 2135 |
+
"grad_norm": 6.457112789154053,
|
| 2136 |
+
"learning_rate": 2.544514646754739e-06,
|
| 2137 |
+
"loss": 0.328,
|
| 2138 |
+
"step": 3040
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 1.7518667432510053,
|
| 2142 |
+
"grad_norm": 11.173613548278809,
|
| 2143 |
+
"learning_rate": 2.4870763928776568e-06,
|
| 2144 |
+
"loss": 0.2807,
|
| 2145 |
+
"step": 3050
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 1.7576105686387133,
|
| 2149 |
+
"grad_norm": 7.816757678985596,
|
| 2150 |
+
"learning_rate": 2.4296381390005747e-06,
|
| 2151 |
+
"loss": 0.3215,
|
| 2152 |
+
"step": 3060
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 1.7633543940264216,
|
| 2156 |
+
"grad_norm": 9.915975570678711,
|
| 2157 |
+
"learning_rate": 2.372199885123492e-06,
|
| 2158 |
+
"loss": 0.4274,
|
| 2159 |
+
"step": 3070
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 1.7690982194141298,
|
| 2163 |
+
"grad_norm": 4.84187650680542,
|
| 2164 |
+
"learning_rate": 2.31476163124641e-06,
|
| 2165 |
+
"loss": 0.2679,
|
| 2166 |
+
"step": 3080
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 1.774842044801838,
|
| 2170 |
+
"grad_norm": 7.7299909591674805,
|
| 2171 |
+
"learning_rate": 2.2573233773693283e-06,
|
| 2172 |
+
"loss": 0.37,
|
| 2173 |
+
"step": 3090
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 1.7805858701895463,
|
| 2177 |
+
"grad_norm": 6.248807430267334,
|
| 2178 |
+
"learning_rate": 2.1998851234922462e-06,
|
| 2179 |
+
"loss": 0.3418,
|
| 2180 |
+
"step": 3100
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 1.7863296955772543,
|
| 2184 |
+
"grad_norm": 7.012147426605225,
|
| 2185 |
+
"learning_rate": 2.1424468696151637e-06,
|
| 2186 |
+
"loss": 0.3138,
|
| 2187 |
+
"step": 3110
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 1.7920735209649625,
|
| 2191 |
+
"grad_norm": 4.839092254638672,
|
| 2192 |
+
"learning_rate": 2.0850086157380816e-06,
|
| 2193 |
+
"loss": 0.3464,
|
| 2194 |
+
"step": 3120
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 1.7978173463526708,
|
| 2198 |
+
"grad_norm": 9.893400192260742,
|
| 2199 |
+
"learning_rate": 2.0275703618609995e-06,
|
| 2200 |
+
"loss": 0.315,
|
| 2201 |
+
"step": 3130
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 1.803561171740379,
|
| 2205 |
+
"grad_norm": 4.713288307189941,
|
| 2206 |
+
"learning_rate": 1.9701321079839174e-06,
|
| 2207 |
+
"loss": 0.2864,
|
| 2208 |
+
"step": 3140
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 1.8093049971280872,
|
| 2212 |
+
"grad_norm": 7.380845546722412,
|
| 2213 |
+
"learning_rate": 1.9126938541068353e-06,
|
| 2214 |
+
"loss": 0.3599,
|
| 2215 |
+
"step": 3150
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 1.8150488225157955,
|
| 2219 |
+
"grad_norm": 8.279121398925781,
|
| 2220 |
+
"learning_rate": 1.8552556002297532e-06,
|
| 2221 |
+
"loss": 0.2871,
|
| 2222 |
+
"step": 3160
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 1.8207926479035037,
|
| 2226 |
+
"grad_norm": 8.517400741577148,
|
| 2227 |
+
"learning_rate": 1.797817346352671e-06,
|
| 2228 |
+
"loss": 0.2782,
|
| 2229 |
+
"step": 3170
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 1.826536473291212,
|
| 2233 |
+
"grad_norm": 8.836138725280762,
|
| 2234 |
+
"learning_rate": 1.740379092475589e-06,
|
| 2235 |
+
"loss": 0.3053,
|
| 2236 |
+
"step": 3180
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 1.8322802986789202,
|
| 2240 |
+
"grad_norm": 8.175244331359863,
|
| 2241 |
+
"learning_rate": 1.6829408385985068e-06,
|
| 2242 |
+
"loss": 0.2729,
|
| 2243 |
+
"step": 3190
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 1.8380241240666284,
|
| 2247 |
+
"grad_norm": 8.452549934387207,
|
| 2248 |
+
"learning_rate": 1.6255025847214245e-06,
|
| 2249 |
+
"loss": 0.3369,
|
| 2250 |
+
"step": 3200
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 1.8437679494543366,
|
| 2254 |
+
"grad_norm": 4.472502708435059,
|
| 2255 |
+
"learning_rate": 1.5680643308443424e-06,
|
| 2256 |
+
"loss": 0.3386,
|
| 2257 |
+
"step": 3210
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 1.8495117748420449,
|
| 2261 |
+
"grad_norm": 6.361071586608887,
|
| 2262 |
+
"learning_rate": 1.5106260769672603e-06,
|
| 2263 |
+
"loss": 0.2744,
|
| 2264 |
+
"step": 3220
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 1.855255600229753,
|
| 2268 |
+
"grad_norm": 5.243939399719238,
|
| 2269 |
+
"learning_rate": 1.4531878230901784e-06,
|
| 2270 |
+
"loss": 0.263,
|
| 2271 |
+
"step": 3230
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 1.8609994256174613,
|
| 2275 |
+
"grad_norm": 11.582174301147461,
|
| 2276 |
+
"learning_rate": 1.3957495692130959e-06,
|
| 2277 |
+
"loss": 0.4315,
|
| 2278 |
+
"step": 3240
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 1.8667432510051696,
|
| 2282 |
+
"grad_norm": 6.288567543029785,
|
| 2283 |
+
"learning_rate": 1.338311315336014e-06,
|
| 2284 |
+
"loss": 0.339,
|
| 2285 |
+
"step": 3250
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 1.8724870763928778,
|
| 2289 |
+
"grad_norm": 12.038402557373047,
|
| 2290 |
+
"learning_rate": 1.2808730614589319e-06,
|
| 2291 |
+
"loss": 0.2851,
|
| 2292 |
+
"step": 3260
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 1.8782309017805858,
|
| 2296 |
+
"grad_norm": 6.310272216796875,
|
| 2297 |
+
"learning_rate": 1.2234348075818495e-06,
|
| 2298 |
+
"loss": 0.2127,
|
| 2299 |
+
"step": 3270
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 1.883974727168294,
|
| 2303 |
+
"grad_norm": 7.386268615722656,
|
| 2304 |
+
"learning_rate": 1.1659965537047674e-06,
|
| 2305 |
+
"loss": 0.3338,
|
| 2306 |
+
"step": 3280
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 1.8897185525560023,
|
| 2310 |
+
"grad_norm": 7.40879487991333,
|
| 2311 |
+
"learning_rate": 1.1085582998276853e-06,
|
| 2312 |
+
"loss": 0.2658,
|
| 2313 |
+
"step": 3290
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 1.8954623779437105,
|
| 2317 |
+
"grad_norm": 13.936086654663086,
|
| 2318 |
+
"learning_rate": 1.0511200459506032e-06,
|
| 2319 |
+
"loss": 0.3539,
|
| 2320 |
+
"step": 3300
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 1.9012062033314188,
|
| 2324 |
+
"grad_norm": 5.246440410614014,
|
| 2325 |
+
"learning_rate": 9.936817920735211e-07,
|
| 2326 |
+
"loss": 0.3869,
|
| 2327 |
+
"step": 3310
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 1.9069500287191268,
|
| 2331 |
+
"grad_norm": 7.692013740539551,
|
| 2332 |
+
"learning_rate": 9.36243538196439e-07,
|
| 2333 |
+
"loss": 0.2657,
|
| 2334 |
+
"step": 3320
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 1.912693854106835,
|
| 2338 |
+
"grad_norm": 10.840241432189941,
|
| 2339 |
+
"learning_rate": 8.788052843193568e-07,
|
| 2340 |
+
"loss": 0.3165,
|
| 2341 |
+
"step": 3330
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 1.9184376794945432,
|
| 2345 |
+
"grad_norm": 5.2672810554504395,
|
| 2346 |
+
"learning_rate": 8.213670304422747e-07,
|
| 2347 |
+
"loss": 0.2489,
|
| 2348 |
+
"step": 3340
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 1.9241815048822515,
|
| 2352 |
+
"grad_norm": 3.8065874576568604,
|
| 2353 |
+
"learning_rate": 7.639287765651925e-07,
|
| 2354 |
+
"loss": 0.3271,
|
| 2355 |
+
"step": 3350
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 1.9299253302699597,
|
| 2359 |
+
"grad_norm": 10.850006103515625,
|
| 2360 |
+
"learning_rate": 7.064905226881104e-07,
|
| 2361 |
+
"loss": 0.3887,
|
| 2362 |
+
"step": 3360
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 1.935669155657668,
|
| 2366 |
+
"grad_norm": 4.718282699584961,
|
| 2367 |
+
"learning_rate": 6.490522688110281e-07,
|
| 2368 |
+
"loss": 0.3125,
|
| 2369 |
+
"step": 3370
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 1.9414129810453762,
|
| 2373 |
+
"grad_norm": 5.016386032104492,
|
| 2374 |
+
"learning_rate": 5.91614014933946e-07,
|
| 2375 |
+
"loss": 0.3018,
|
| 2376 |
+
"step": 3380
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 1.9471568064330844,
|
| 2380 |
+
"grad_norm": 11.608333587646484,
|
| 2381 |
+
"learning_rate": 5.341757610568639e-07,
|
| 2382 |
+
"loss": 0.2911,
|
| 2383 |
+
"step": 3390
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 1.9529006318207927,
|
| 2387 |
+
"grad_norm": 7.1915459632873535,
|
| 2388 |
+
"learning_rate": 4.7673750717978177e-07,
|
| 2389 |
+
"loss": 0.2974,
|
| 2390 |
+
"step": 3400
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 1.9586444572085009,
|
| 2394 |
+
"grad_norm": 9.097732543945312,
|
| 2395 |
+
"learning_rate": 4.192992533026996e-07,
|
| 2396 |
+
"loss": 0.3127,
|
| 2397 |
+
"step": 3410
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 1.9643882825962091,
|
| 2401 |
+
"grad_norm": 5.897636890411377,
|
| 2402 |
+
"learning_rate": 3.6186099942561744e-07,
|
| 2403 |
+
"loss": 0.3128,
|
| 2404 |
+
"step": 3420
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 1.9701321079839174,
|
| 2408 |
+
"grad_norm": 5.362806797027588,
|
| 2409 |
+
"learning_rate": 3.0442274554853533e-07,
|
| 2410 |
+
"loss": 0.2742,
|
| 2411 |
+
"step": 3430
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 1.9758759333716256,
|
| 2415 |
+
"grad_norm": 8.615423202514648,
|
| 2416 |
+
"learning_rate": 2.469844916714532e-07,
|
| 2417 |
+
"loss": 0.3787,
|
| 2418 |
+
"step": 3440
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 1.9816197587593338,
|
| 2422 |
+
"grad_norm": 8.756539344787598,
|
| 2423 |
+
"learning_rate": 1.8954623779437106e-07,
|
| 2424 |
+
"loss": 0.3705,
|
| 2425 |
+
"step": 3450
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 1.987363584147042,
|
| 2429 |
+
"grad_norm": 11.84693717956543,
|
| 2430 |
+
"learning_rate": 1.3210798391728893e-07,
|
| 2431 |
+
"loss": 0.3027,
|
| 2432 |
+
"step": 3460
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 1.9931074095347503,
|
| 2436 |
+
"grad_norm": 11.413233757019043,
|
| 2437 |
+
"learning_rate": 7.466973004020678e-08,
|
| 2438 |
+
"loss": 0.3549,
|
| 2439 |
+
"step": 3470
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 1.9988512349224583,
|
| 2443 |
+
"grad_norm": 5.79984188079834,
|
| 2444 |
+
"learning_rate": 1.7231476163124642e-08,
|
| 2445 |
+
"loss": 0.3109,
|
| 2446 |
+
"step": 3480
|
| 2447 |
+
}
|
| 2448 |
+
],
|
| 2449 |
+
"logging_steps": 10,
|
| 2450 |
+
"max_steps": 3482,
|
| 2451 |
+
"num_input_tokens_seen": 0,
|
| 2452 |
+
"num_train_epochs": 2,
|
| 2453 |
+
"save_steps": 500,
|
| 2454 |
+
"stateful_callbacks": {
|
| 2455 |
+
"TrainerControl": {
|
| 2456 |
+
"args": {
|
| 2457 |
+
"should_epoch_stop": false,
|
| 2458 |
+
"should_evaluate": false,
|
| 2459 |
+
"should_log": false,
|
| 2460 |
+
"should_save": true,
|
| 2461 |
+
"should_training_stop": true
|
| 2462 |
+
},
|
| 2463 |
+
"attributes": {}
|
| 2464 |
+
}
|
| 2465 |
+
},
|
| 2466 |
+
"total_flos": 7376844496355328.0,
|
| 2467 |
+
"train_batch_size": 16,
|
| 2468 |
+
"trial_name": null,
|
| 2469 |
+
"trial_params": null
|
| 2470 |
+
}
|
checkpoint-3482/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7e7cb6431ba79d66c2c9c534739c31af31901b4f69b86ef7cd6fa47d58097e4a
|
| 3 |
+
size 5368
|
checkpoint-3482/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
config.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation": "gelu",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"DistilBertForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.1,
|
| 7 |
+
"dim": 768,
|
| 8 |
+
"dropout": 0.3,
|
| 9 |
+
"dtype": "float32",
|
| 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 |
+
"transformers_version": "4.57.1",
|
| 23 |
+
"vocab_size": 30522
|
| 24 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:44a9699f5eae1c12be68d7d6783b3db0d0136ec5a126bd85d7435845fcefef10
|
| 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
|
|
|