Commit
•
acc6eb0
1
Parent(s):
2a06e7e
Training in progress, step 1032, checkpoint
Browse files- last-checkpoint/config.json +22 -0
- last-checkpoint/model.safetensors +3 -0
- last-checkpoint/optimizer.pt +3 -0
- last-checkpoint/rng_state_0.pth +3 -0
- last-checkpoint/rng_state_1.pth +3 -0
- last-checkpoint/scheduler.pt +3 -0
- last-checkpoint/special_tokens_map.json +7 -0
- last-checkpoint/tokenizer.json +0 -0
- last-checkpoint/tokenizer_config.json +55 -0
- last-checkpoint/trainer_state.json +1827 -0
- last-checkpoint/training_args.bin +3 -0
- last-checkpoint/vocab.txt +0 -0
last-checkpoint/config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation": "gelu",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"DistilBertForMaskedLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.1,
|
| 7 |
+
"dim": 768,
|
| 8 |
+
"dropout": 0.1,
|
| 9 |
+
"hidden_dim": 3072,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"max_position_embeddings": 1024,
|
| 12 |
+
"model_type": "distilbert",
|
| 13 |
+
"n_heads": 12,
|
| 14 |
+
"n_layers": 12,
|
| 15 |
+
"pad_token_id": 0,
|
| 16 |
+
"qa_dropout": 0.1,
|
| 17 |
+
"seq_classif_dropout": 0.2,
|
| 18 |
+
"sinusoidal_pos_embds": false,
|
| 19 |
+
"torch_dtype": "float32",
|
| 20 |
+
"transformers_version": "4.39.0.dev0",
|
| 21 |
+
"vocab_size": 30522
|
| 22 |
+
}
|
last-checkpoint/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8c89b353fa09e3ce83e793b919d4f980f3146c0a40010d2f8e71ce59bc6caf97
|
| 3 |
+
size 439648328
|
last-checkpoint/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:344c11e8655bdf0d6d795b4c0210797f8ec26fa495ff00315eb79210095b95ba
|
| 3 |
+
size 879415866
|
last-checkpoint/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:483cd0d3b3d771ead4322faf51b93c007348b3287d9725b7e7e67b944f15f025
|
| 3 |
+
size 14512
|
last-checkpoint/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c92a2c39191c76f8bc730f293bc42dc1167804104de7d5d46ecbdd03b9c6e8d2
|
| 3 |
+
size 14512
|
last-checkpoint/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3100131d4aa200ed28289bfc07f3c3092bfa6c034debb09f60179e7579429d9e
|
| 3 |
+
size 1064
|
last-checkpoint/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 |
+
}
|
last-checkpoint/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
last-checkpoint/tokenizer_config.json
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": true,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_lower_case": true,
|
| 47 |
+
"mask_token": "[MASK]",
|
| 48 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 49 |
+
"pad_token": "[PAD]",
|
| 50 |
+
"sep_token": "[SEP]",
|
| 51 |
+
"strip_accents": null,
|
| 52 |
+
"tokenize_chinese_chars": true,
|
| 53 |
+
"tokenizer_class": "DistilBertTokenizer",
|
| 54 |
+
"unk_token": "[UNK]"
|
| 55 |
+
}
|
last-checkpoint/trainer_state.json
ADDED
|
@@ -0,0 +1,1827 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.10005817335660268,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1032,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0,
|
| 13 |
+
"grad_norm": 3.656641960144043,
|
| 14 |
+
"learning_rate": 4.998060888113244e-05,
|
| 15 |
+
"loss": 9.8019,
|
| 16 |
+
"step": 4
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0,
|
| 20 |
+
"grad_norm": 3.3737926483154297,
|
| 21 |
+
"learning_rate": 4.9961217762264883e-05,
|
| 22 |
+
"loss": 9.12,
|
| 23 |
+
"step": 8
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0,
|
| 27 |
+
"grad_norm": 3.5734024047851562,
|
| 28 |
+
"learning_rate": 4.9941826643397324e-05,
|
| 29 |
+
"loss": 8.8002,
|
| 30 |
+
"step": 12
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0,
|
| 34 |
+
"grad_norm": 3.0836071968078613,
|
| 35 |
+
"learning_rate": 4.992243552452977e-05,
|
| 36 |
+
"loss": 8.5483,
|
| 37 |
+
"step": 16
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0,
|
| 41 |
+
"grad_norm": 3.126896619796753,
|
| 42 |
+
"learning_rate": 4.9903044405662205e-05,
|
| 43 |
+
"loss": 8.376,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.0,
|
| 48 |
+
"grad_norm": 2.795313835144043,
|
| 49 |
+
"learning_rate": 4.988365328679465e-05,
|
| 50 |
+
"loss": 8.2286,
|
| 51 |
+
"step": 24
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.0,
|
| 55 |
+
"grad_norm": 2.847642183303833,
|
| 56 |
+
"learning_rate": 4.9864262167927086e-05,
|
| 57 |
+
"loss": 8.0428,
|
| 58 |
+
"step": 28
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.0,
|
| 62 |
+
"grad_norm": 2.6489531993865967,
|
| 63 |
+
"learning_rate": 4.9844871049059533e-05,
|
| 64 |
+
"loss": 7.8429,
|
| 65 |
+
"step": 32
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.0,
|
| 69 |
+
"grad_norm": 2.569993495941162,
|
| 70 |
+
"learning_rate": 4.9825479930191974e-05,
|
| 71 |
+
"loss": 7.6876,
|
| 72 |
+
"step": 36
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.0,
|
| 76 |
+
"grad_norm": 2.3502109050750732,
|
| 77 |
+
"learning_rate": 4.9806088811324415e-05,
|
| 78 |
+
"loss": 7.5277,
|
| 79 |
+
"step": 40
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.0,
|
| 83 |
+
"grad_norm": 2.2252399921417236,
|
| 84 |
+
"learning_rate": 4.9786697692456855e-05,
|
| 85 |
+
"loss": 7.3008,
|
| 86 |
+
"step": 44
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.0,
|
| 90 |
+
"grad_norm": 2.2069296836853027,
|
| 91 |
+
"learning_rate": 4.97673065735893e-05,
|
| 92 |
+
"loss": 7.1233,
|
| 93 |
+
"step": 48
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.01,
|
| 97 |
+
"grad_norm": 2.133172035217285,
|
| 98 |
+
"learning_rate": 4.9747915454721736e-05,
|
| 99 |
+
"loss": 6.9925,
|
| 100 |
+
"step": 52
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.01,
|
| 104 |
+
"grad_norm": 1.830748438835144,
|
| 105 |
+
"learning_rate": 4.9728524335854183e-05,
|
| 106 |
+
"loss": 6.8808,
|
| 107 |
+
"step": 56
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.01,
|
| 111 |
+
"grad_norm": 1.7571477890014648,
|
| 112 |
+
"learning_rate": 4.970913321698662e-05,
|
| 113 |
+
"loss": 6.7121,
|
| 114 |
+
"step": 60
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.01,
|
| 118 |
+
"grad_norm": 1.6238043308258057,
|
| 119 |
+
"learning_rate": 4.9689742098119065e-05,
|
| 120 |
+
"loss": 6.6879,
|
| 121 |
+
"step": 64
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.01,
|
| 125 |
+
"grad_norm": 1.460631012916565,
|
| 126 |
+
"learning_rate": 4.9670350979251505e-05,
|
| 127 |
+
"loss": 6.4892,
|
| 128 |
+
"step": 68
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.01,
|
| 132 |
+
"grad_norm": 1.399665355682373,
|
| 133 |
+
"learning_rate": 4.9650959860383946e-05,
|
| 134 |
+
"loss": 6.4278,
|
| 135 |
+
"step": 72
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.01,
|
| 139 |
+
"grad_norm": 1.2306325435638428,
|
| 140 |
+
"learning_rate": 4.9631568741516386e-05,
|
| 141 |
+
"loss": 6.43,
|
| 142 |
+
"step": 76
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.01,
|
| 146 |
+
"grad_norm": 1.1850312948226929,
|
| 147 |
+
"learning_rate": 4.9612177622648833e-05,
|
| 148 |
+
"loss": 6.3628,
|
| 149 |
+
"step": 80
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.01,
|
| 153 |
+
"grad_norm": 1.3839452266693115,
|
| 154 |
+
"learning_rate": 4.959278650378127e-05,
|
| 155 |
+
"loss": 6.1944,
|
| 156 |
+
"step": 84
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.01,
|
| 160 |
+
"grad_norm": 1.2195193767547607,
|
| 161 |
+
"learning_rate": 4.9573395384913715e-05,
|
| 162 |
+
"loss": 6.2424,
|
| 163 |
+
"step": 88
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.01,
|
| 167 |
+
"grad_norm": 1.1323491334915161,
|
| 168 |
+
"learning_rate": 4.9554004266046155e-05,
|
| 169 |
+
"loss": 6.112,
|
| 170 |
+
"step": 92
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.01,
|
| 174 |
+
"grad_norm": 1.2749617099761963,
|
| 175 |
+
"learning_rate": 4.9534613147178596e-05,
|
| 176 |
+
"loss": 6.13,
|
| 177 |
+
"step": 96
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.01,
|
| 181 |
+
"grad_norm": 1.2026777267456055,
|
| 182 |
+
"learning_rate": 4.9515222028311036e-05,
|
| 183 |
+
"loss": 6.0626,
|
| 184 |
+
"step": 100
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.01,
|
| 188 |
+
"grad_norm": 1.4500185251235962,
|
| 189 |
+
"learning_rate": 4.949583090944348e-05,
|
| 190 |
+
"loss": 6.0918,
|
| 191 |
+
"step": 104
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.01,
|
| 195 |
+
"grad_norm": 1.2848747968673706,
|
| 196 |
+
"learning_rate": 4.947643979057592e-05,
|
| 197 |
+
"loss": 6.1688,
|
| 198 |
+
"step": 108
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.01,
|
| 202 |
+
"grad_norm": 1.3187015056610107,
|
| 203 |
+
"learning_rate": 4.9457048671708365e-05,
|
| 204 |
+
"loss": 6.0728,
|
| 205 |
+
"step": 112
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.01,
|
| 209 |
+
"grad_norm": 1.487337589263916,
|
| 210 |
+
"learning_rate": 4.94376575528408e-05,
|
| 211 |
+
"loss": 6.0801,
|
| 212 |
+
"step": 116
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.01,
|
| 216 |
+
"grad_norm": 1.191052794456482,
|
| 217 |
+
"learning_rate": 4.9418266433973246e-05,
|
| 218 |
+
"loss": 6.1361,
|
| 219 |
+
"step": 120
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.01,
|
| 223 |
+
"grad_norm": 1.5774741172790527,
|
| 224 |
+
"learning_rate": 4.9398875315105686e-05,
|
| 225 |
+
"loss": 6.055,
|
| 226 |
+
"step": 124
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.01,
|
| 230 |
+
"grad_norm": 1.0442208051681519,
|
| 231 |
+
"learning_rate": 4.937948419623813e-05,
|
| 232 |
+
"loss": 6.0708,
|
| 233 |
+
"step": 128
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.01,
|
| 237 |
+
"grad_norm": 1.4346356391906738,
|
| 238 |
+
"learning_rate": 4.936009307737057e-05,
|
| 239 |
+
"loss": 6.0169,
|
| 240 |
+
"step": 132
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.01,
|
| 244 |
+
"grad_norm": 1.2307249307632446,
|
| 245 |
+
"learning_rate": 4.934070195850301e-05,
|
| 246 |
+
"loss": 6.0547,
|
| 247 |
+
"step": 136
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.01,
|
| 251 |
+
"grad_norm": 1.1539560556411743,
|
| 252 |
+
"learning_rate": 4.932131083963545e-05,
|
| 253 |
+
"loss": 6.0166,
|
| 254 |
+
"step": 140
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.01,
|
| 258 |
+
"grad_norm": 1.1614229679107666,
|
| 259 |
+
"learning_rate": 4.930191972076789e-05,
|
| 260 |
+
"loss": 6.1866,
|
| 261 |
+
"step": 144
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.01,
|
| 265 |
+
"grad_norm": 1.0596864223480225,
|
| 266 |
+
"learning_rate": 4.928252860190033e-05,
|
| 267 |
+
"loss": 5.9715,
|
| 268 |
+
"step": 148
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.01,
|
| 272 |
+
"grad_norm": 1.111221194267273,
|
| 273 |
+
"learning_rate": 4.926313748303277e-05,
|
| 274 |
+
"loss": 6.0272,
|
| 275 |
+
"step": 152
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.02,
|
| 279 |
+
"grad_norm": 0.9217305183410645,
|
| 280 |
+
"learning_rate": 4.924374636416522e-05,
|
| 281 |
+
"loss": 6.0135,
|
| 282 |
+
"step": 156
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.02,
|
| 286 |
+
"grad_norm": 0.968235969543457,
|
| 287 |
+
"learning_rate": 4.922435524529765e-05,
|
| 288 |
+
"loss": 5.9969,
|
| 289 |
+
"step": 160
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.02,
|
| 293 |
+
"grad_norm": 1.1550750732421875,
|
| 294 |
+
"learning_rate": 4.92049641264301e-05,
|
| 295 |
+
"loss": 6.0566,
|
| 296 |
+
"step": 164
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.02,
|
| 300 |
+
"grad_norm": 1.5655795335769653,
|
| 301 |
+
"learning_rate": 4.918557300756254e-05,
|
| 302 |
+
"loss": 5.9619,
|
| 303 |
+
"step": 168
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.02,
|
| 307 |
+
"grad_norm": 1.2823691368103027,
|
| 308 |
+
"learning_rate": 4.916618188869498e-05,
|
| 309 |
+
"loss": 5.876,
|
| 310 |
+
"step": 172
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.02,
|
| 314 |
+
"grad_norm": 0.913962721824646,
|
| 315 |
+
"learning_rate": 4.914679076982742e-05,
|
| 316 |
+
"loss": 6.0023,
|
| 317 |
+
"step": 176
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.02,
|
| 321 |
+
"grad_norm": 1.1342990398406982,
|
| 322 |
+
"learning_rate": 4.912739965095986e-05,
|
| 323 |
+
"loss": 5.9118,
|
| 324 |
+
"step": 180
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.02,
|
| 328 |
+
"grad_norm": 1.1372032165527344,
|
| 329 |
+
"learning_rate": 4.91080085320923e-05,
|
| 330 |
+
"loss": 5.9543,
|
| 331 |
+
"step": 184
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.02,
|
| 335 |
+
"grad_norm": 1.1001873016357422,
|
| 336 |
+
"learning_rate": 4.908861741322475e-05,
|
| 337 |
+
"loss": 6.0478,
|
| 338 |
+
"step": 188
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.02,
|
| 342 |
+
"grad_norm": 1.163060188293457,
|
| 343 |
+
"learning_rate": 4.906922629435718e-05,
|
| 344 |
+
"loss": 5.9354,
|
| 345 |
+
"step": 192
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.02,
|
| 349 |
+
"grad_norm": 1.022900938987732,
|
| 350 |
+
"learning_rate": 4.904983517548963e-05,
|
| 351 |
+
"loss": 5.8227,
|
| 352 |
+
"step": 196
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.02,
|
| 356 |
+
"grad_norm": 1.2557566165924072,
|
| 357 |
+
"learning_rate": 4.903044405662207e-05,
|
| 358 |
+
"loss": 5.898,
|
| 359 |
+
"step": 200
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.02,
|
| 363 |
+
"grad_norm": 1.260396122932434,
|
| 364 |
+
"learning_rate": 4.901105293775451e-05,
|
| 365 |
+
"loss": 5.9742,
|
| 366 |
+
"step": 204
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.02,
|
| 370 |
+
"grad_norm": 1.0883902311325073,
|
| 371 |
+
"learning_rate": 4.899166181888695e-05,
|
| 372 |
+
"loss": 5.8642,
|
| 373 |
+
"step": 208
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.02,
|
| 377 |
+
"grad_norm": 1.07965886592865,
|
| 378 |
+
"learning_rate": 4.89722707000194e-05,
|
| 379 |
+
"loss": 5.9836,
|
| 380 |
+
"step": 212
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.02,
|
| 384 |
+
"grad_norm": 1.12454092502594,
|
| 385 |
+
"learning_rate": 4.895287958115183e-05,
|
| 386 |
+
"loss": 5.9383,
|
| 387 |
+
"step": 216
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.02,
|
| 391 |
+
"grad_norm": 1.1025103330612183,
|
| 392 |
+
"learning_rate": 4.893348846228428e-05,
|
| 393 |
+
"loss": 5.8433,
|
| 394 |
+
"step": 220
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.02,
|
| 398 |
+
"grad_norm": 1.1046481132507324,
|
| 399 |
+
"learning_rate": 4.891409734341671e-05,
|
| 400 |
+
"loss": 5.8649,
|
| 401 |
+
"step": 224
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.02,
|
| 405 |
+
"grad_norm": 1.356872320175171,
|
| 406 |
+
"learning_rate": 4.889470622454916e-05,
|
| 407 |
+
"loss": 5.9069,
|
| 408 |
+
"step": 228
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.02,
|
| 412 |
+
"grad_norm": 1.1604045629501343,
|
| 413 |
+
"learning_rate": 4.88753151056816e-05,
|
| 414 |
+
"loss": 5.8704,
|
| 415 |
+
"step": 232
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.02,
|
| 419 |
+
"grad_norm": 1.1360890865325928,
|
| 420 |
+
"learning_rate": 4.885592398681404e-05,
|
| 421 |
+
"loss": 5.8723,
|
| 422 |
+
"step": 236
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.02,
|
| 426 |
+
"grad_norm": 0.8874178528785706,
|
| 427 |
+
"learning_rate": 4.883653286794648e-05,
|
| 428 |
+
"loss": 5.8548,
|
| 429 |
+
"step": 240
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.02,
|
| 433 |
+
"grad_norm": 1.0104761123657227,
|
| 434 |
+
"learning_rate": 4.881714174907893e-05,
|
| 435 |
+
"loss": 5.8931,
|
| 436 |
+
"step": 244
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.02,
|
| 440 |
+
"grad_norm": 1.1234338283538818,
|
| 441 |
+
"learning_rate": 4.879775063021136e-05,
|
| 442 |
+
"loss": 5.8915,
|
| 443 |
+
"step": 248
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.02,
|
| 447 |
+
"grad_norm": 1.2352111339569092,
|
| 448 |
+
"learning_rate": 4.877835951134381e-05,
|
| 449 |
+
"loss": 5.9084,
|
| 450 |
+
"step": 252
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.02,
|
| 454 |
+
"grad_norm": 0.9969356656074524,
|
| 455 |
+
"learning_rate": 4.875896839247625e-05,
|
| 456 |
+
"loss": 5.9788,
|
| 457 |
+
"step": 256
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.03,
|
| 461 |
+
"grad_norm": 1.4352229833602905,
|
| 462 |
+
"learning_rate": 4.873957727360869e-05,
|
| 463 |
+
"loss": 5.8121,
|
| 464 |
+
"step": 260
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.03,
|
| 468 |
+
"grad_norm": 1.3081138134002686,
|
| 469 |
+
"learning_rate": 4.872018615474113e-05,
|
| 470 |
+
"loss": 5.8714,
|
| 471 |
+
"step": 264
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.03,
|
| 475 |
+
"grad_norm": 1.1553027629852295,
|
| 476 |
+
"learning_rate": 4.870079503587357e-05,
|
| 477 |
+
"loss": 5.9062,
|
| 478 |
+
"step": 268
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.03,
|
| 482 |
+
"grad_norm": 1.328942060470581,
|
| 483 |
+
"learning_rate": 4.868140391700601e-05,
|
| 484 |
+
"loss": 5.9659,
|
| 485 |
+
"step": 272
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.03,
|
| 489 |
+
"grad_norm": 1.092142939567566,
|
| 490 |
+
"learning_rate": 4.8662012798138454e-05,
|
| 491 |
+
"loss": 5.9436,
|
| 492 |
+
"step": 276
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.03,
|
| 496 |
+
"grad_norm": 0.8468914031982422,
|
| 497 |
+
"learning_rate": 4.8642621679270894e-05,
|
| 498 |
+
"loss": 5.9176,
|
| 499 |
+
"step": 280
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.03,
|
| 503 |
+
"grad_norm": 1.1835949420928955,
|
| 504 |
+
"learning_rate": 4.8623230560403335e-05,
|
| 505 |
+
"loss": 5.8968,
|
| 506 |
+
"step": 284
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.03,
|
| 510 |
+
"grad_norm": 1.401023507118225,
|
| 511 |
+
"learning_rate": 4.860383944153578e-05,
|
| 512 |
+
"loss": 5.8126,
|
| 513 |
+
"step": 288
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.03,
|
| 517 |
+
"grad_norm": 1.2995976209640503,
|
| 518 |
+
"learning_rate": 4.8584448322668216e-05,
|
| 519 |
+
"loss": 5.9769,
|
| 520 |
+
"step": 292
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.03,
|
| 524 |
+
"grad_norm": 1.0936148166656494,
|
| 525 |
+
"learning_rate": 4.856505720380066e-05,
|
| 526 |
+
"loss": 5.8341,
|
| 527 |
+
"step": 296
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.03,
|
| 531 |
+
"grad_norm": 1.121054768562317,
|
| 532 |
+
"learning_rate": 4.85456660849331e-05,
|
| 533 |
+
"loss": 5.8082,
|
| 534 |
+
"step": 300
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.03,
|
| 538 |
+
"grad_norm": 1.011224389076233,
|
| 539 |
+
"learning_rate": 4.8526274966065544e-05,
|
| 540 |
+
"loss": 5.8437,
|
| 541 |
+
"step": 304
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.03,
|
| 545 |
+
"grad_norm": 0.909989058971405,
|
| 546 |
+
"learning_rate": 4.8506883847197985e-05,
|
| 547 |
+
"loss": 5.8217,
|
| 548 |
+
"step": 308
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.03,
|
| 552 |
+
"grad_norm": 0.9056807160377502,
|
| 553 |
+
"learning_rate": 4.8487492728330425e-05,
|
| 554 |
+
"loss": 5.8791,
|
| 555 |
+
"step": 312
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.03,
|
| 559 |
+
"grad_norm": 0.990393340587616,
|
| 560 |
+
"learning_rate": 4.8468101609462866e-05,
|
| 561 |
+
"loss": 5.9895,
|
| 562 |
+
"step": 316
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.03,
|
| 566 |
+
"grad_norm": 1.252443790435791,
|
| 567 |
+
"learning_rate": 4.844871049059531e-05,
|
| 568 |
+
"loss": 5.9232,
|
| 569 |
+
"step": 320
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.03,
|
| 573 |
+
"grad_norm": 0.9842864274978638,
|
| 574 |
+
"learning_rate": 4.842931937172775e-05,
|
| 575 |
+
"loss": 5.8068,
|
| 576 |
+
"step": 324
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.03,
|
| 580 |
+
"grad_norm": 1.050066351890564,
|
| 581 |
+
"learning_rate": 4.8409928252860194e-05,
|
| 582 |
+
"loss": 5.8112,
|
| 583 |
+
"step": 328
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.03,
|
| 587 |
+
"grad_norm": 0.9735043048858643,
|
| 588 |
+
"learning_rate": 4.8390537133992635e-05,
|
| 589 |
+
"loss": 5.8329,
|
| 590 |
+
"step": 332
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.03,
|
| 594 |
+
"grad_norm": 0.948833703994751,
|
| 595 |
+
"learning_rate": 4.8371146015125075e-05,
|
| 596 |
+
"loss": 5.8564,
|
| 597 |
+
"step": 336
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.03,
|
| 601 |
+
"grad_norm": 1.0226508378982544,
|
| 602 |
+
"learning_rate": 4.8351754896257516e-05,
|
| 603 |
+
"loss": 5.7493,
|
| 604 |
+
"step": 340
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.03,
|
| 608 |
+
"grad_norm": 1.09911048412323,
|
| 609 |
+
"learning_rate": 4.8332363777389956e-05,
|
| 610 |
+
"loss": 5.7015,
|
| 611 |
+
"step": 344
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.03,
|
| 615 |
+
"grad_norm": 1.1777935028076172,
|
| 616 |
+
"learning_rate": 4.83129726585224e-05,
|
| 617 |
+
"loss": 5.7534,
|
| 618 |
+
"step": 348
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.03,
|
| 622 |
+
"grad_norm": 1.1616647243499756,
|
| 623 |
+
"learning_rate": 4.8293581539654844e-05,
|
| 624 |
+
"loss": 5.8506,
|
| 625 |
+
"step": 352
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.03,
|
| 629 |
+
"grad_norm": 1.0601903200149536,
|
| 630 |
+
"learning_rate": 4.827419042078728e-05,
|
| 631 |
+
"loss": 5.8917,
|
| 632 |
+
"step": 356
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.03,
|
| 636 |
+
"grad_norm": 0.9961209893226624,
|
| 637 |
+
"learning_rate": 4.8254799301919725e-05,
|
| 638 |
+
"loss": 5.797,
|
| 639 |
+
"step": 360
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.04,
|
| 643 |
+
"grad_norm": 1.0286229848861694,
|
| 644 |
+
"learning_rate": 4.8235408183052166e-05,
|
| 645 |
+
"loss": 5.888,
|
| 646 |
+
"step": 364
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.04,
|
| 650 |
+
"grad_norm": 1.0672661066055298,
|
| 651 |
+
"learning_rate": 4.8216017064184606e-05,
|
| 652 |
+
"loss": 5.823,
|
| 653 |
+
"step": 368
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.04,
|
| 657 |
+
"grad_norm": 0.9099803566932678,
|
| 658 |
+
"learning_rate": 4.819662594531705e-05,
|
| 659 |
+
"loss": 5.7411,
|
| 660 |
+
"step": 372
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.04,
|
| 664 |
+
"grad_norm": 1.1643503904342651,
|
| 665 |
+
"learning_rate": 4.8177234826449494e-05,
|
| 666 |
+
"loss": 5.9417,
|
| 667 |
+
"step": 376
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.04,
|
| 671 |
+
"grad_norm": 0.9987436532974243,
|
| 672 |
+
"learning_rate": 4.815784370758193e-05,
|
| 673 |
+
"loss": 5.8176,
|
| 674 |
+
"step": 380
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.04,
|
| 678 |
+
"grad_norm": 1.1946054697036743,
|
| 679 |
+
"learning_rate": 4.8138452588714375e-05,
|
| 680 |
+
"loss": 5.9005,
|
| 681 |
+
"step": 384
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.04,
|
| 685 |
+
"grad_norm": 0.9317137002944946,
|
| 686 |
+
"learning_rate": 4.811906146984681e-05,
|
| 687 |
+
"loss": 5.8017,
|
| 688 |
+
"step": 388
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.04,
|
| 692 |
+
"grad_norm": 1.0581262111663818,
|
| 693 |
+
"learning_rate": 4.8099670350979256e-05,
|
| 694 |
+
"loss": 5.794,
|
| 695 |
+
"step": 392
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.04,
|
| 699 |
+
"grad_norm": 1.1224080324172974,
|
| 700 |
+
"learning_rate": 4.80802792321117e-05,
|
| 701 |
+
"loss": 5.8161,
|
| 702 |
+
"step": 396
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.04,
|
| 706 |
+
"grad_norm": 1.2604461908340454,
|
| 707 |
+
"learning_rate": 4.806088811324414e-05,
|
| 708 |
+
"loss": 5.8109,
|
| 709 |
+
"step": 400
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.04,
|
| 713 |
+
"grad_norm": 1.1275701522827148,
|
| 714 |
+
"learning_rate": 4.804149699437658e-05,
|
| 715 |
+
"loss": 5.8104,
|
| 716 |
+
"step": 404
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.04,
|
| 720 |
+
"grad_norm": 1.012634038925171,
|
| 721 |
+
"learning_rate": 4.802210587550902e-05,
|
| 722 |
+
"loss": 5.8702,
|
| 723 |
+
"step": 408
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.04,
|
| 727 |
+
"grad_norm": 0.8785901665687561,
|
| 728 |
+
"learning_rate": 4.800271475664146e-05,
|
| 729 |
+
"loss": 5.8498,
|
| 730 |
+
"step": 412
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.04,
|
| 734 |
+
"grad_norm": 0.8808034062385559,
|
| 735 |
+
"learning_rate": 4.79833236377739e-05,
|
| 736 |
+
"loss": 5.7474,
|
| 737 |
+
"step": 416
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.04,
|
| 741 |
+
"grad_norm": 1.1132124662399292,
|
| 742 |
+
"learning_rate": 4.796393251890634e-05,
|
| 743 |
+
"loss": 5.7462,
|
| 744 |
+
"step": 420
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.04,
|
| 748 |
+
"grad_norm": 0.9468419551849365,
|
| 749 |
+
"learning_rate": 4.794454140003878e-05,
|
| 750 |
+
"loss": 5.7767,
|
| 751 |
+
"step": 424
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.04,
|
| 755 |
+
"grad_norm": 0.9365801215171814,
|
| 756 |
+
"learning_rate": 4.792515028117123e-05,
|
| 757 |
+
"loss": 5.8072,
|
| 758 |
+
"step": 428
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.04,
|
| 762 |
+
"grad_norm": 1.0467591285705566,
|
| 763 |
+
"learning_rate": 4.790575916230366e-05,
|
| 764 |
+
"loss": 5.7773,
|
| 765 |
+
"step": 432
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.04,
|
| 769 |
+
"grad_norm": 1.1843376159667969,
|
| 770 |
+
"learning_rate": 4.788636804343611e-05,
|
| 771 |
+
"loss": 5.9357,
|
| 772 |
+
"step": 436
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.04,
|
| 776 |
+
"grad_norm": 1.139872431755066,
|
| 777 |
+
"learning_rate": 4.786697692456855e-05,
|
| 778 |
+
"loss": 5.7744,
|
| 779 |
+
"step": 440
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.04,
|
| 783 |
+
"grad_norm": 0.9096530079841614,
|
| 784 |
+
"learning_rate": 4.784758580570099e-05,
|
| 785 |
+
"loss": 5.9122,
|
| 786 |
+
"step": 444
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.04,
|
| 790 |
+
"grad_norm": 1.0682241916656494,
|
| 791 |
+
"learning_rate": 4.782819468683343e-05,
|
| 792 |
+
"loss": 5.8587,
|
| 793 |
+
"step": 448
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.04,
|
| 797 |
+
"grad_norm": 0.99188232421875,
|
| 798 |
+
"learning_rate": 4.780880356796588e-05,
|
| 799 |
+
"loss": 5.8096,
|
| 800 |
+
"step": 452
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.04,
|
| 804 |
+
"grad_norm": 1.0947283506393433,
|
| 805 |
+
"learning_rate": 4.778941244909831e-05,
|
| 806 |
+
"loss": 5.6552,
|
| 807 |
+
"step": 456
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.04,
|
| 811 |
+
"grad_norm": 1.0802547931671143,
|
| 812 |
+
"learning_rate": 4.777002133023076e-05,
|
| 813 |
+
"loss": 5.9245,
|
| 814 |
+
"step": 460
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.04,
|
| 818 |
+
"grad_norm": 1.164602518081665,
|
| 819 |
+
"learning_rate": 4.775063021136319e-05,
|
| 820 |
+
"loss": 5.8362,
|
| 821 |
+
"step": 464
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.05,
|
| 825 |
+
"grad_norm": 1.1340216398239136,
|
| 826 |
+
"learning_rate": 4.773123909249564e-05,
|
| 827 |
+
"loss": 5.8477,
|
| 828 |
+
"step": 468
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.05,
|
| 832 |
+
"grad_norm": 1.0476304292678833,
|
| 833 |
+
"learning_rate": 4.771184797362808e-05,
|
| 834 |
+
"loss": 5.8336,
|
| 835 |
+
"step": 472
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.05,
|
| 839 |
+
"grad_norm": 0.9962639808654785,
|
| 840 |
+
"learning_rate": 4.769245685476052e-05,
|
| 841 |
+
"loss": 5.7876,
|
| 842 |
+
"step": 476
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.05,
|
| 846 |
+
"grad_norm": 0.9998891949653625,
|
| 847 |
+
"learning_rate": 4.767306573589296e-05,
|
| 848 |
+
"loss": 5.6897,
|
| 849 |
+
"step": 480
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.05,
|
| 853 |
+
"grad_norm": 1.054436445236206,
|
| 854 |
+
"learning_rate": 4.765367461702541e-05,
|
| 855 |
+
"loss": 5.6461,
|
| 856 |
+
"step": 484
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.05,
|
| 860 |
+
"grad_norm": 0.9963980913162231,
|
| 861 |
+
"learning_rate": 4.763428349815784e-05,
|
| 862 |
+
"loss": 5.7391,
|
| 863 |
+
"step": 488
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.05,
|
| 867 |
+
"grad_norm": 1.077802062034607,
|
| 868 |
+
"learning_rate": 4.761489237929029e-05,
|
| 869 |
+
"loss": 5.772,
|
| 870 |
+
"step": 492
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.05,
|
| 874 |
+
"grad_norm": 1.3317667245864868,
|
| 875 |
+
"learning_rate": 4.759550126042273e-05,
|
| 876 |
+
"loss": 5.7657,
|
| 877 |
+
"step": 496
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.05,
|
| 881 |
+
"grad_norm": 1.028090000152588,
|
| 882 |
+
"learning_rate": 4.757611014155517e-05,
|
| 883 |
+
"loss": 5.7417,
|
| 884 |
+
"step": 500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.05,
|
| 888 |
+
"grad_norm": 0.9719061255455017,
|
| 889 |
+
"learning_rate": 4.755671902268761e-05,
|
| 890 |
+
"loss": 5.8042,
|
| 891 |
+
"step": 504
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.05,
|
| 895 |
+
"grad_norm": 1.067749261856079,
|
| 896 |
+
"learning_rate": 4.753732790382005e-05,
|
| 897 |
+
"loss": 5.7923,
|
| 898 |
+
"step": 508
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.05,
|
| 902 |
+
"grad_norm": 0.9830943942070007,
|
| 903 |
+
"learning_rate": 4.751793678495249e-05,
|
| 904 |
+
"loss": 5.7911,
|
| 905 |
+
"step": 512
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.05,
|
| 909 |
+
"grad_norm": 1.0226807594299316,
|
| 910 |
+
"learning_rate": 4.749854566608494e-05,
|
| 911 |
+
"loss": 5.7696,
|
| 912 |
+
"step": 516
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.05,
|
| 916 |
+
"grad_norm": 1.3311954736709595,
|
| 917 |
+
"learning_rate": 4.7479154547217374e-05,
|
| 918 |
+
"loss": 5.8689,
|
| 919 |
+
"step": 520
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.05,
|
| 923 |
+
"grad_norm": 0.9809809923171997,
|
| 924 |
+
"learning_rate": 4.745976342834982e-05,
|
| 925 |
+
"loss": 5.8023,
|
| 926 |
+
"step": 524
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.05,
|
| 930 |
+
"grad_norm": 1.1488884687423706,
|
| 931 |
+
"learning_rate": 4.744037230948226e-05,
|
| 932 |
+
"loss": 5.7116,
|
| 933 |
+
"step": 528
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.05,
|
| 937 |
+
"grad_norm": 1.0571842193603516,
|
| 938 |
+
"learning_rate": 4.74209811906147e-05,
|
| 939 |
+
"loss": 5.7105,
|
| 940 |
+
"step": 532
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.05,
|
| 944 |
+
"grad_norm": 0.9185091257095337,
|
| 945 |
+
"learning_rate": 4.740159007174714e-05,
|
| 946 |
+
"loss": 5.8675,
|
| 947 |
+
"step": 536
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.05,
|
| 951 |
+
"grad_norm": 0.9164991974830627,
|
| 952 |
+
"learning_rate": 4.738219895287958e-05,
|
| 953 |
+
"loss": 5.7231,
|
| 954 |
+
"step": 540
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.05,
|
| 958 |
+
"grad_norm": 1.0493993759155273,
|
| 959 |
+
"learning_rate": 4.7362807834012024e-05,
|
| 960 |
+
"loss": 5.8437,
|
| 961 |
+
"step": 544
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.05,
|
| 965 |
+
"grad_norm": 0.8653802275657654,
|
| 966 |
+
"learning_rate": 4.7343416715144464e-05,
|
| 967 |
+
"loss": 5.652,
|
| 968 |
+
"step": 548
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.05,
|
| 972 |
+
"grad_norm": 0.975098192691803,
|
| 973 |
+
"learning_rate": 4.7324025596276905e-05,
|
| 974 |
+
"loss": 5.7353,
|
| 975 |
+
"step": 552
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.05,
|
| 979 |
+
"grad_norm": 0.9267313480377197,
|
| 980 |
+
"learning_rate": 4.7304634477409345e-05,
|
| 981 |
+
"loss": 5.7886,
|
| 982 |
+
"step": 556
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.05,
|
| 986 |
+
"grad_norm": 0.9583150148391724,
|
| 987 |
+
"learning_rate": 4.728524335854179e-05,
|
| 988 |
+
"loss": 5.6864,
|
| 989 |
+
"step": 560
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.05,
|
| 993 |
+
"grad_norm": 0.9664693474769592,
|
| 994 |
+
"learning_rate": 4.7265852239674227e-05,
|
| 995 |
+
"loss": 5.8314,
|
| 996 |
+
"step": 564
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.06,
|
| 1000 |
+
"grad_norm": 0.8994282484054565,
|
| 1001 |
+
"learning_rate": 4.7246461120806674e-05,
|
| 1002 |
+
"loss": 5.7865,
|
| 1003 |
+
"step": 568
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.06,
|
| 1007 |
+
"grad_norm": 1.0095504522323608,
|
| 1008 |
+
"learning_rate": 4.7227070001939114e-05,
|
| 1009 |
+
"loss": 5.7627,
|
| 1010 |
+
"step": 572
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.06,
|
| 1014 |
+
"grad_norm": 1.0120139122009277,
|
| 1015 |
+
"learning_rate": 4.7207678883071555e-05,
|
| 1016 |
+
"loss": 5.6769,
|
| 1017 |
+
"step": 576
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.06,
|
| 1021 |
+
"grad_norm": 0.9483413100242615,
|
| 1022 |
+
"learning_rate": 4.7188287764203995e-05,
|
| 1023 |
+
"loss": 5.7615,
|
| 1024 |
+
"step": 580
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.06,
|
| 1028 |
+
"grad_norm": 0.9208195209503174,
|
| 1029 |
+
"learning_rate": 4.7168896645336436e-05,
|
| 1030 |
+
"loss": 5.7178,
|
| 1031 |
+
"step": 584
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.06,
|
| 1035 |
+
"grad_norm": 1.0568493604660034,
|
| 1036 |
+
"learning_rate": 4.7149505526468876e-05,
|
| 1037 |
+
"loss": 5.6305,
|
| 1038 |
+
"step": 588
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.06,
|
| 1042 |
+
"grad_norm": 0.9036898016929626,
|
| 1043 |
+
"learning_rate": 4.7130114407601324e-05,
|
| 1044 |
+
"loss": 5.7088,
|
| 1045 |
+
"step": 592
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.06,
|
| 1049 |
+
"grad_norm": 1.0219190120697021,
|
| 1050 |
+
"learning_rate": 4.711072328873376e-05,
|
| 1051 |
+
"loss": 5.7359,
|
| 1052 |
+
"step": 596
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.06,
|
| 1056 |
+
"grad_norm": 1.1148089170455933,
|
| 1057 |
+
"learning_rate": 4.7091332169866205e-05,
|
| 1058 |
+
"loss": 5.7653,
|
| 1059 |
+
"step": 600
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.06,
|
| 1063 |
+
"grad_norm": 1.0402499437332153,
|
| 1064 |
+
"learning_rate": 4.7071941050998645e-05,
|
| 1065 |
+
"loss": 5.7291,
|
| 1066 |
+
"step": 604
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.06,
|
| 1070 |
+
"grad_norm": 1.2846848964691162,
|
| 1071 |
+
"learning_rate": 4.7052549932131086e-05,
|
| 1072 |
+
"loss": 5.8432,
|
| 1073 |
+
"step": 608
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.06,
|
| 1077 |
+
"grad_norm": 1.0562716722488403,
|
| 1078 |
+
"learning_rate": 4.7033158813263526e-05,
|
| 1079 |
+
"loss": 5.7242,
|
| 1080 |
+
"step": 612
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.06,
|
| 1084 |
+
"grad_norm": 0.9270595908164978,
|
| 1085 |
+
"learning_rate": 4.7013767694395974e-05,
|
| 1086 |
+
"loss": 5.735,
|
| 1087 |
+
"step": 616
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.06,
|
| 1091 |
+
"grad_norm": 0.9591754674911499,
|
| 1092 |
+
"learning_rate": 4.699437657552841e-05,
|
| 1093 |
+
"loss": 5.6973,
|
| 1094 |
+
"step": 620
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.06,
|
| 1098 |
+
"grad_norm": 1.020099401473999,
|
| 1099 |
+
"learning_rate": 4.6974985456660855e-05,
|
| 1100 |
+
"loss": 5.7211,
|
| 1101 |
+
"step": 624
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.06,
|
| 1105 |
+
"grad_norm": 0.9728394150733948,
|
| 1106 |
+
"learning_rate": 4.695559433779329e-05,
|
| 1107 |
+
"loss": 5.7558,
|
| 1108 |
+
"step": 628
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.06,
|
| 1112 |
+
"grad_norm": 0.9725939631462097,
|
| 1113 |
+
"learning_rate": 4.6936203218925736e-05,
|
| 1114 |
+
"loss": 5.8508,
|
| 1115 |
+
"step": 632
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.06,
|
| 1119 |
+
"grad_norm": 1.0429859161376953,
|
| 1120 |
+
"learning_rate": 4.6916812100058176e-05,
|
| 1121 |
+
"loss": 5.7557,
|
| 1122 |
+
"step": 636
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.06,
|
| 1126 |
+
"grad_norm": 1.1550756692886353,
|
| 1127 |
+
"learning_rate": 4.689742098119062e-05,
|
| 1128 |
+
"loss": 5.7864,
|
| 1129 |
+
"step": 640
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.06,
|
| 1133 |
+
"grad_norm": 1.277512788772583,
|
| 1134 |
+
"learning_rate": 4.687802986232306e-05,
|
| 1135 |
+
"loss": 5.731,
|
| 1136 |
+
"step": 644
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.06,
|
| 1140 |
+
"grad_norm": 1.170222282409668,
|
| 1141 |
+
"learning_rate": 4.6858638743455505e-05,
|
| 1142 |
+
"loss": 5.6716,
|
| 1143 |
+
"step": 648
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.06,
|
| 1147 |
+
"grad_norm": 1.0293292999267578,
|
| 1148 |
+
"learning_rate": 4.683924762458794e-05,
|
| 1149 |
+
"loss": 5.7762,
|
| 1150 |
+
"step": 652
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.06,
|
| 1154 |
+
"grad_norm": 1.0924861431121826,
|
| 1155 |
+
"learning_rate": 4.6819856505720386e-05,
|
| 1156 |
+
"loss": 5.7223,
|
| 1157 |
+
"step": 656
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.06,
|
| 1161 |
+
"grad_norm": 0.949212908744812,
|
| 1162 |
+
"learning_rate": 4.680046538685282e-05,
|
| 1163 |
+
"loss": 5.7425,
|
| 1164 |
+
"step": 660
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.06,
|
| 1168 |
+
"grad_norm": 1.0186933279037476,
|
| 1169 |
+
"learning_rate": 4.678107426798527e-05,
|
| 1170 |
+
"loss": 5.8361,
|
| 1171 |
+
"step": 664
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.06,
|
| 1175 |
+
"grad_norm": 1.0421589612960815,
|
| 1176 |
+
"learning_rate": 4.676168314911771e-05,
|
| 1177 |
+
"loss": 5.7093,
|
| 1178 |
+
"step": 668
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.07,
|
| 1182 |
+
"grad_norm": 1.061271071434021,
|
| 1183 |
+
"learning_rate": 4.674229203025015e-05,
|
| 1184 |
+
"loss": 5.6382,
|
| 1185 |
+
"step": 672
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.07,
|
| 1189 |
+
"grad_norm": 1.0228848457336426,
|
| 1190 |
+
"learning_rate": 4.672290091138259e-05,
|
| 1191 |
+
"loss": 5.6845,
|
| 1192 |
+
"step": 676
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.07,
|
| 1196 |
+
"grad_norm": 1.0488477945327759,
|
| 1197 |
+
"learning_rate": 4.670350979251503e-05,
|
| 1198 |
+
"loss": 5.6403,
|
| 1199 |
+
"step": 680
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.07,
|
| 1203 |
+
"grad_norm": 0.9394490718841553,
|
| 1204 |
+
"learning_rate": 4.668411867364747e-05,
|
| 1205 |
+
"loss": 5.6733,
|
| 1206 |
+
"step": 684
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.07,
|
| 1210 |
+
"grad_norm": 1.140890121459961,
|
| 1211 |
+
"learning_rate": 4.666472755477991e-05,
|
| 1212 |
+
"loss": 5.7066,
|
| 1213 |
+
"step": 688
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.07,
|
| 1217 |
+
"grad_norm": 0.9432203769683838,
|
| 1218 |
+
"learning_rate": 4.664533643591236e-05,
|
| 1219 |
+
"loss": 5.7846,
|
| 1220 |
+
"step": 692
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.07,
|
| 1224 |
+
"grad_norm": 1.0256586074829102,
|
| 1225 |
+
"learning_rate": 4.662594531704479e-05,
|
| 1226 |
+
"loss": 5.766,
|
| 1227 |
+
"step": 696
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.07,
|
| 1231 |
+
"grad_norm": 1.0383509397506714,
|
| 1232 |
+
"learning_rate": 4.660655419817724e-05,
|
| 1233 |
+
"loss": 5.8138,
|
| 1234 |
+
"step": 700
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.07,
|
| 1238 |
+
"grad_norm": 0.9286032319068909,
|
| 1239 |
+
"learning_rate": 4.658716307930968e-05,
|
| 1240 |
+
"loss": 5.7233,
|
| 1241 |
+
"step": 704
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.07,
|
| 1245 |
+
"grad_norm": 0.9339187145233154,
|
| 1246 |
+
"learning_rate": 4.656777196044212e-05,
|
| 1247 |
+
"loss": 5.803,
|
| 1248 |
+
"step": 708
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.07,
|
| 1252 |
+
"grad_norm": 0.9535828828811646,
|
| 1253 |
+
"learning_rate": 4.654838084157456e-05,
|
| 1254 |
+
"loss": 5.8002,
|
| 1255 |
+
"step": 712
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.07,
|
| 1259 |
+
"grad_norm": 1.2881035804748535,
|
| 1260 |
+
"learning_rate": 4.6528989722707e-05,
|
| 1261 |
+
"loss": 5.5501,
|
| 1262 |
+
"step": 716
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.07,
|
| 1266 |
+
"grad_norm": 0.9991912841796875,
|
| 1267 |
+
"learning_rate": 4.650959860383944e-05,
|
| 1268 |
+
"loss": 5.7732,
|
| 1269 |
+
"step": 720
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.07,
|
| 1273 |
+
"grad_norm": 1.071502923965454,
|
| 1274 |
+
"learning_rate": 4.649020748497189e-05,
|
| 1275 |
+
"loss": 5.6494,
|
| 1276 |
+
"step": 724
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.07,
|
| 1280 |
+
"grad_norm": 0.9422402381896973,
|
| 1281 |
+
"learning_rate": 4.647081636610432e-05,
|
| 1282 |
+
"loss": 5.7539,
|
| 1283 |
+
"step": 728
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.07,
|
| 1287 |
+
"grad_norm": 1.2175371646881104,
|
| 1288 |
+
"learning_rate": 4.645142524723677e-05,
|
| 1289 |
+
"loss": 5.6747,
|
| 1290 |
+
"step": 732
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.07,
|
| 1294 |
+
"grad_norm": 1.1459147930145264,
|
| 1295 |
+
"learning_rate": 4.643203412836921e-05,
|
| 1296 |
+
"loss": 5.6895,
|
| 1297 |
+
"step": 736
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.07,
|
| 1301 |
+
"grad_norm": 0.9659698009490967,
|
| 1302 |
+
"learning_rate": 4.641264300950165e-05,
|
| 1303 |
+
"loss": 5.6517,
|
| 1304 |
+
"step": 740
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.07,
|
| 1308 |
+
"grad_norm": 1.071323275566101,
|
| 1309 |
+
"learning_rate": 4.639325189063409e-05,
|
| 1310 |
+
"loss": 5.653,
|
| 1311 |
+
"step": 744
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.07,
|
| 1315 |
+
"grad_norm": 1.1298364400863647,
|
| 1316 |
+
"learning_rate": 4.637386077176653e-05,
|
| 1317 |
+
"loss": 5.6975,
|
| 1318 |
+
"step": 748
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.07,
|
| 1322 |
+
"grad_norm": 0.9889796376228333,
|
| 1323 |
+
"learning_rate": 4.635446965289897e-05,
|
| 1324 |
+
"loss": 5.8007,
|
| 1325 |
+
"step": 752
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.07,
|
| 1329 |
+
"grad_norm": 0.9490829110145569,
|
| 1330 |
+
"learning_rate": 4.633507853403142e-05,
|
| 1331 |
+
"loss": 5.6969,
|
| 1332 |
+
"step": 756
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.07,
|
| 1336 |
+
"grad_norm": 1.0920511484146118,
|
| 1337 |
+
"learning_rate": 4.6315687415163853e-05,
|
| 1338 |
+
"loss": 5.74,
|
| 1339 |
+
"step": 760
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.07,
|
| 1343 |
+
"grad_norm": 1.039368748664856,
|
| 1344 |
+
"learning_rate": 4.62962962962963e-05,
|
| 1345 |
+
"loss": 5.7002,
|
| 1346 |
+
"step": 764
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.07,
|
| 1350 |
+
"grad_norm": 0.9749051928520203,
|
| 1351 |
+
"learning_rate": 4.627690517742874e-05,
|
| 1352 |
+
"loss": 5.6879,
|
| 1353 |
+
"step": 768
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.07,
|
| 1357 |
+
"grad_norm": 0.9410939812660217,
|
| 1358 |
+
"learning_rate": 4.625751405856118e-05,
|
| 1359 |
+
"loss": 5.7138,
|
| 1360 |
+
"step": 772
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.08,
|
| 1364 |
+
"grad_norm": 0.9284390807151794,
|
| 1365 |
+
"learning_rate": 4.623812293969362e-05,
|
| 1366 |
+
"loss": 5.6772,
|
| 1367 |
+
"step": 776
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.08,
|
| 1371 |
+
"grad_norm": 1.220140814781189,
|
| 1372 |
+
"learning_rate": 4.621873182082606e-05,
|
| 1373 |
+
"loss": 5.6187,
|
| 1374 |
+
"step": 780
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.08,
|
| 1378 |
+
"grad_norm": 0.9217903017997742,
|
| 1379 |
+
"learning_rate": 4.6199340701958503e-05,
|
| 1380 |
+
"loss": 5.7051,
|
| 1381 |
+
"step": 784
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.08,
|
| 1385 |
+
"grad_norm": 0.9127422571182251,
|
| 1386 |
+
"learning_rate": 4.617994958309095e-05,
|
| 1387 |
+
"loss": 5.6821,
|
| 1388 |
+
"step": 788
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.08,
|
| 1392 |
+
"grad_norm": 0.9815553426742554,
|
| 1393 |
+
"learning_rate": 4.6160558464223385e-05,
|
| 1394 |
+
"loss": 5.7365,
|
| 1395 |
+
"step": 792
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.08,
|
| 1399 |
+
"grad_norm": 0.8972638249397278,
|
| 1400 |
+
"learning_rate": 4.614116734535583e-05,
|
| 1401 |
+
"loss": 5.6835,
|
| 1402 |
+
"step": 796
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.08,
|
| 1406 |
+
"grad_norm": 0.9297725558280945,
|
| 1407 |
+
"learning_rate": 4.612177622648827e-05,
|
| 1408 |
+
"loss": 5.6304,
|
| 1409 |
+
"step": 800
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.08,
|
| 1413 |
+
"grad_norm": 0.9928802847862244,
|
| 1414 |
+
"learning_rate": 4.610238510762071e-05,
|
| 1415 |
+
"loss": 5.8121,
|
| 1416 |
+
"step": 804
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.08,
|
| 1420 |
+
"grad_norm": 0.9002383351325989,
|
| 1421 |
+
"learning_rate": 4.6082993988753153e-05,
|
| 1422 |
+
"loss": 5.6457,
|
| 1423 |
+
"step": 808
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.08,
|
| 1427 |
+
"grad_norm": 0.8902143239974976,
|
| 1428 |
+
"learning_rate": 4.6063602869885594e-05,
|
| 1429 |
+
"loss": 5.7276,
|
| 1430 |
+
"step": 812
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.08,
|
| 1434 |
+
"grad_norm": 1.069921612739563,
|
| 1435 |
+
"learning_rate": 4.6044211751018035e-05,
|
| 1436 |
+
"loss": 5.6827,
|
| 1437 |
+
"step": 816
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.08,
|
| 1441 |
+
"grad_norm": 0.9564573764801025,
|
| 1442 |
+
"learning_rate": 4.6024820632150475e-05,
|
| 1443 |
+
"loss": 5.694,
|
| 1444 |
+
"step": 820
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.08,
|
| 1448 |
+
"grad_norm": 0.9404289126396179,
|
| 1449 |
+
"learning_rate": 4.6005429513282916e-05,
|
| 1450 |
+
"loss": 5.585,
|
| 1451 |
+
"step": 824
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.08,
|
| 1455 |
+
"grad_norm": 0.9606718420982361,
|
| 1456 |
+
"learning_rate": 4.5986038394415356e-05,
|
| 1457 |
+
"loss": 5.7258,
|
| 1458 |
+
"step": 828
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.08,
|
| 1462 |
+
"grad_norm": 0.893195629119873,
|
| 1463 |
+
"learning_rate": 4.5966647275547803e-05,
|
| 1464 |
+
"loss": 5.6816,
|
| 1465 |
+
"step": 832
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.08,
|
| 1469 |
+
"grad_norm": 0.9530772566795349,
|
| 1470 |
+
"learning_rate": 4.5947256156680244e-05,
|
| 1471 |
+
"loss": 5.7424,
|
| 1472 |
+
"step": 836
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.08,
|
| 1476 |
+
"grad_norm": 1.1925373077392578,
|
| 1477 |
+
"learning_rate": 4.5927865037812685e-05,
|
| 1478 |
+
"loss": 5.6958,
|
| 1479 |
+
"step": 840
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.08,
|
| 1483 |
+
"grad_norm": 1.0818183422088623,
|
| 1484 |
+
"learning_rate": 4.5908473918945125e-05,
|
| 1485 |
+
"loss": 5.6273,
|
| 1486 |
+
"step": 844
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.08,
|
| 1490 |
+
"grad_norm": 1.1746087074279785,
|
| 1491 |
+
"learning_rate": 4.5889082800077566e-05,
|
| 1492 |
+
"loss": 5.6699,
|
| 1493 |
+
"step": 848
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.08,
|
| 1497 |
+
"grad_norm": 1.069509744644165,
|
| 1498 |
+
"learning_rate": 4.5869691681210006e-05,
|
| 1499 |
+
"loss": 5.677,
|
| 1500 |
+
"step": 852
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.08,
|
| 1504 |
+
"grad_norm": 1.0898882150650024,
|
| 1505 |
+
"learning_rate": 4.5850300562342453e-05,
|
| 1506 |
+
"loss": 5.723,
|
| 1507 |
+
"step": 856
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.08,
|
| 1511 |
+
"grad_norm": 0.9786805510520935,
|
| 1512 |
+
"learning_rate": 4.583090944347489e-05,
|
| 1513 |
+
"loss": 5.7109,
|
| 1514 |
+
"step": 860
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.08,
|
| 1518 |
+
"grad_norm": 1.0218449831008911,
|
| 1519 |
+
"learning_rate": 4.5811518324607335e-05,
|
| 1520 |
+
"loss": 5.7927,
|
| 1521 |
+
"step": 864
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.08,
|
| 1525 |
+
"grad_norm": 1.0597094297409058,
|
| 1526 |
+
"learning_rate": 4.579212720573977e-05,
|
| 1527 |
+
"loss": 5.6456,
|
| 1528 |
+
"step": 868
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.08,
|
| 1532 |
+
"grad_norm": 1.015222430229187,
|
| 1533 |
+
"learning_rate": 4.5772736086872216e-05,
|
| 1534 |
+
"loss": 5.7023,
|
| 1535 |
+
"step": 872
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.08,
|
| 1539 |
+
"grad_norm": 1.0093826055526733,
|
| 1540 |
+
"learning_rate": 4.5753344968004656e-05,
|
| 1541 |
+
"loss": 5.673,
|
| 1542 |
+
"step": 876
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.09,
|
| 1546 |
+
"grad_norm": 1.1618988513946533,
|
| 1547 |
+
"learning_rate": 4.57339538491371e-05,
|
| 1548 |
+
"loss": 5.5492,
|
| 1549 |
+
"step": 880
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.09,
|
| 1553 |
+
"grad_norm": 0.9958293437957764,
|
| 1554 |
+
"learning_rate": 4.571456273026954e-05,
|
| 1555 |
+
"loss": 5.5945,
|
| 1556 |
+
"step": 884
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.09,
|
| 1560 |
+
"grad_norm": 0.9605300426483154,
|
| 1561 |
+
"learning_rate": 4.5695171611401985e-05,
|
| 1562 |
+
"loss": 5.6439,
|
| 1563 |
+
"step": 888
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.09,
|
| 1567 |
+
"grad_norm": 1.0881812572479248,
|
| 1568 |
+
"learning_rate": 4.567578049253442e-05,
|
| 1569 |
+
"loss": 5.7346,
|
| 1570 |
+
"step": 892
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.09,
|
| 1574 |
+
"grad_norm": 1.1047258377075195,
|
| 1575 |
+
"learning_rate": 4.5656389373666866e-05,
|
| 1576 |
+
"loss": 5.6754,
|
| 1577 |
+
"step": 896
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.09,
|
| 1581 |
+
"grad_norm": 1.0597559213638306,
|
| 1582 |
+
"learning_rate": 4.56369982547993e-05,
|
| 1583 |
+
"loss": 5.6655,
|
| 1584 |
+
"step": 900
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.09,
|
| 1588 |
+
"grad_norm": 1.0080434083938599,
|
| 1589 |
+
"learning_rate": 4.561760713593175e-05,
|
| 1590 |
+
"loss": 5.637,
|
| 1591 |
+
"step": 904
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.09,
|
| 1595 |
+
"grad_norm": 1.1175477504730225,
|
| 1596 |
+
"learning_rate": 4.559821601706419e-05,
|
| 1597 |
+
"loss": 5.6114,
|
| 1598 |
+
"step": 908
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.09,
|
| 1602 |
+
"grad_norm": 1.309441089630127,
|
| 1603 |
+
"learning_rate": 4.557882489819663e-05,
|
| 1604 |
+
"loss": 5.64,
|
| 1605 |
+
"step": 912
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.09,
|
| 1609 |
+
"grad_norm": 1.0713026523590088,
|
| 1610 |
+
"learning_rate": 4.555943377932907e-05,
|
| 1611 |
+
"loss": 5.6433,
|
| 1612 |
+
"step": 916
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.09,
|
| 1616 |
+
"grad_norm": 1.1929994821548462,
|
| 1617 |
+
"learning_rate": 4.5540042660461516e-05,
|
| 1618 |
+
"loss": 5.6503,
|
| 1619 |
+
"step": 920
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.09,
|
| 1623 |
+
"grad_norm": 1.0933693647384644,
|
| 1624 |
+
"learning_rate": 4.552065154159395e-05,
|
| 1625 |
+
"loss": 5.68,
|
| 1626 |
+
"step": 924
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.09,
|
| 1630 |
+
"grad_norm": 0.9823185801506042,
|
| 1631 |
+
"learning_rate": 4.55012604227264e-05,
|
| 1632 |
+
"loss": 5.6941,
|
| 1633 |
+
"step": 928
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.09,
|
| 1637 |
+
"grad_norm": 0.9883749485015869,
|
| 1638 |
+
"learning_rate": 4.548186930385884e-05,
|
| 1639 |
+
"loss": 5.5973,
|
| 1640 |
+
"step": 932
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.09,
|
| 1644 |
+
"grad_norm": 0.9289098381996155,
|
| 1645 |
+
"learning_rate": 4.546247818499128e-05,
|
| 1646 |
+
"loss": 5.7444,
|
| 1647 |
+
"step": 936
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.09,
|
| 1651 |
+
"grad_norm": 1.0262410640716553,
|
| 1652 |
+
"learning_rate": 4.544308706612372e-05,
|
| 1653 |
+
"loss": 5.6235,
|
| 1654 |
+
"step": 940
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.09,
|
| 1658 |
+
"grad_norm": 1.0129894018173218,
|
| 1659 |
+
"learning_rate": 4.542369594725616e-05,
|
| 1660 |
+
"loss": 5.6469,
|
| 1661 |
+
"step": 944
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.09,
|
| 1665 |
+
"grad_norm": 1.0500363111495972,
|
| 1666 |
+
"learning_rate": 4.54043048283886e-05,
|
| 1667 |
+
"loss": 5.6281,
|
| 1668 |
+
"step": 948
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.09,
|
| 1672 |
+
"grad_norm": 0.9239895343780518,
|
| 1673 |
+
"learning_rate": 4.538491370952104e-05,
|
| 1674 |
+
"loss": 5.6151,
|
| 1675 |
+
"step": 952
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.09,
|
| 1679 |
+
"grad_norm": 0.9726962447166443,
|
| 1680 |
+
"learning_rate": 4.536552259065348e-05,
|
| 1681 |
+
"loss": 5.6146,
|
| 1682 |
+
"step": 956
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.09,
|
| 1686 |
+
"grad_norm": 0.9822578430175781,
|
| 1687 |
+
"learning_rate": 4.534613147178592e-05,
|
| 1688 |
+
"loss": 5.5852,
|
| 1689 |
+
"step": 960
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.09,
|
| 1693 |
+
"grad_norm": 0.9551820755004883,
|
| 1694 |
+
"learning_rate": 4.532674035291837e-05,
|
| 1695 |
+
"loss": 5.7069,
|
| 1696 |
+
"step": 964
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.09,
|
| 1700 |
+
"grad_norm": 0.9644606709480286,
|
| 1701 |
+
"learning_rate": 4.530734923405081e-05,
|
| 1702 |
+
"loss": 5.6178,
|
| 1703 |
+
"step": 968
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.09,
|
| 1707 |
+
"grad_norm": 1.1014901399612427,
|
| 1708 |
+
"learning_rate": 4.528795811518325e-05,
|
| 1709 |
+
"loss": 5.6131,
|
| 1710 |
+
"step": 972
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.09,
|
| 1714 |
+
"grad_norm": 1.0390726327896118,
|
| 1715 |
+
"learning_rate": 4.526856699631569e-05,
|
| 1716 |
+
"loss": 5.5601,
|
| 1717 |
+
"step": 976
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.1,
|
| 1721 |
+
"grad_norm": 0.9854323863983154,
|
| 1722 |
+
"learning_rate": 4.524917587744813e-05,
|
| 1723 |
+
"loss": 5.6175,
|
| 1724 |
+
"step": 980
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.1,
|
| 1728 |
+
"grad_norm": 0.9083386659622192,
|
| 1729 |
+
"learning_rate": 4.522978475858057e-05,
|
| 1730 |
+
"loss": 5.6013,
|
| 1731 |
+
"step": 984
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.1,
|
| 1735 |
+
"grad_norm": 1.246685266494751,
|
| 1736 |
+
"learning_rate": 4.521039363971301e-05,
|
| 1737 |
+
"loss": 5.7068,
|
| 1738 |
+
"step": 988
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.1,
|
| 1742 |
+
"grad_norm": 1.0527753829956055,
|
| 1743 |
+
"learning_rate": 4.519100252084545e-05,
|
| 1744 |
+
"loss": 5.6358,
|
| 1745 |
+
"step": 992
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.1,
|
| 1749 |
+
"grad_norm": 1.0731735229492188,
|
| 1750 |
+
"learning_rate": 4.51716114019779e-05,
|
| 1751 |
+
"loss": 5.6828,
|
| 1752 |
+
"step": 996
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.1,
|
| 1756 |
+
"grad_norm": 1.0552845001220703,
|
| 1757 |
+
"learning_rate": 4.515222028311033e-05,
|
| 1758 |
+
"loss": 5.6828,
|
| 1759 |
+
"step": 1000
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.1,
|
| 1763 |
+
"grad_norm": 0.990972101688385,
|
| 1764 |
+
"learning_rate": 4.513282916424278e-05,
|
| 1765 |
+
"loss": 5.7003,
|
| 1766 |
+
"step": 1004
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.1,
|
| 1770 |
+
"grad_norm": 1.1035810708999634,
|
| 1771 |
+
"learning_rate": 4.511343804537522e-05,
|
| 1772 |
+
"loss": 5.7133,
|
| 1773 |
+
"step": 1008
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.1,
|
| 1777 |
+
"grad_norm": 1.2226297855377197,
|
| 1778 |
+
"learning_rate": 4.509404692650766e-05,
|
| 1779 |
+
"loss": 5.6895,
|
| 1780 |
+
"step": 1012
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.1,
|
| 1784 |
+
"grad_norm": 1.1234639883041382,
|
| 1785 |
+
"learning_rate": 4.50746558076401e-05,
|
| 1786 |
+
"loss": 5.7874,
|
| 1787 |
+
"step": 1016
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.1,
|
| 1791 |
+
"grad_norm": 1.1130845546722412,
|
| 1792 |
+
"learning_rate": 4.505526468877254e-05,
|
| 1793 |
+
"loss": 5.7322,
|
| 1794 |
+
"step": 1020
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.1,
|
| 1798 |
+
"grad_norm": 0.8997195363044739,
|
| 1799 |
+
"learning_rate": 4.503587356990498e-05,
|
| 1800 |
+
"loss": 5.5914,
|
| 1801 |
+
"step": 1024
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.1,
|
| 1805 |
+
"grad_norm": 0.9451141953468323,
|
| 1806 |
+
"learning_rate": 4.501648245103743e-05,
|
| 1807 |
+
"loss": 5.68,
|
| 1808 |
+
"step": 1028
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.1,
|
| 1812 |
+
"grad_norm": 1.0043590068817139,
|
| 1813 |
+
"learning_rate": 4.4997091332169864e-05,
|
| 1814 |
+
"loss": 5.7082,
|
| 1815 |
+
"step": 1032
|
| 1816 |
+
}
|
| 1817 |
+
],
|
| 1818 |
+
"logging_steps": 4,
|
| 1819 |
+
"max_steps": 10314,
|
| 1820 |
+
"num_input_tokens_seen": 0,
|
| 1821 |
+
"num_train_epochs": 1,
|
| 1822 |
+
"save_steps": 1032,
|
| 1823 |
+
"total_flos": 8692075805540352.0,
|
| 1824 |
+
"train_batch_size": 8,
|
| 1825 |
+
"trial_name": null,
|
| 1826 |
+
"trial_params": null
|
| 1827 |
+
}
|
last-checkpoint/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e7b81f989eff6657e32c531872f9030ee0428a8f966fc8286ce1ae7013c2645c
|
| 3 |
+
size 4984
|
last-checkpoint/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|