Upload 6 files
Browse files- config.json +25 -0
- special_tokens_map.json +37 -0
- tokenizer.json +0 -0
- tokenizer_config.json +56 -0
- trainer_state.json +1476 -0
- vocab.txt +0 -0
config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertModel"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"dtype": "float16",
|
| 8 |
+
"gradient_checkpointing": false,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"layer_norm_eps": 1e-12,
|
| 15 |
+
"max_position_embeddings": 512,
|
| 16 |
+
"model_type": "bert",
|
| 17 |
+
"num_attention_heads": 12,
|
| 18 |
+
"num_hidden_layers": 12,
|
| 19 |
+
"pad_token_id": 0,
|
| 20 |
+
"position_embedding_type": "absolute",
|
| 21 |
+
"transformers_version": "4.57.1",
|
| 22 |
+
"type_vocab_size": 2,
|
| 23 |
+
"use_cache": true,
|
| 24 |
+
"vocab_size": 30522
|
| 25 |
+
}
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": {
|
| 3 |
+
"content": "[CLS]",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"mask_token": {
|
| 10 |
+
"content": "[MASK]",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "[PAD]",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"sep_token": {
|
| 24 |
+
"content": "[SEP]",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"unk_token": {
|
| 31 |
+
"content": "[UNK]",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
}
|
| 37 |
+
}
|
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": true,
|
| 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": "BertTokenizer",
|
| 55 |
+
"unk_token": "[UNK]"
|
| 56 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1476 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 2.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 2060,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.009708737864077669,
|
| 14 |
+
"grad_norm": 8.540091514587402,
|
| 15 |
+
"learning_rate": 3.883495145631068e-07,
|
| 16 |
+
"loss": 2.734,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.019417475728155338,
|
| 21 |
+
"grad_norm": 8.14619255065918,
|
| 22 |
+
"learning_rate": 1.359223300970874e-06,
|
| 23 |
+
"loss": 2.6711,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.02912621359223301,
|
| 28 |
+
"grad_norm": 6.858204364776611,
|
| 29 |
+
"learning_rate": 2.330097087378641e-06,
|
| 30 |
+
"loss": 2.6055,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.038834951456310676,
|
| 35 |
+
"grad_norm": 6.228450298309326,
|
| 36 |
+
"learning_rate": 3.300970873786408e-06,
|
| 37 |
+
"loss": 2.5322,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.04854368932038835,
|
| 42 |
+
"grad_norm": 5.512765407562256,
|
| 43 |
+
"learning_rate": 4.271844660194175e-06,
|
| 44 |
+
"loss": 2.466,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.05825242718446602,
|
| 49 |
+
"grad_norm": 5.114351272583008,
|
| 50 |
+
"learning_rate": 5.242718446601942e-06,
|
| 51 |
+
"loss": 2.4828,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.06796116504854369,
|
| 56 |
+
"grad_norm": 4.749820232391357,
|
| 57 |
+
"learning_rate": 6.213592233009709e-06,
|
| 58 |
+
"loss": 2.4529,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.07766990291262135,
|
| 63 |
+
"grad_norm": 4.962618827819824,
|
| 64 |
+
"learning_rate": 7.184466019417476e-06,
|
| 65 |
+
"loss": 2.3377,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.08737864077669903,
|
| 70 |
+
"grad_norm": 4.963841915130615,
|
| 71 |
+
"learning_rate": 8.155339805825243e-06,
|
| 72 |
+
"loss": 2.3914,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.0970873786407767,
|
| 77 |
+
"grad_norm": 5.020650386810303,
|
| 78 |
+
"learning_rate": 9.12621359223301e-06,
|
| 79 |
+
"loss": 2.2748,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.10679611650485436,
|
| 84 |
+
"grad_norm": 4.788837909698486,
|
| 85 |
+
"learning_rate": 1.0097087378640778e-05,
|
| 86 |
+
"loss": 2.2975,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.11650485436893204,
|
| 91 |
+
"grad_norm": 5.005253791809082,
|
| 92 |
+
"learning_rate": 1.1067961165048544e-05,
|
| 93 |
+
"loss": 2.3025,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.1262135922330097,
|
| 98 |
+
"grad_norm": 5.113918304443359,
|
| 99 |
+
"learning_rate": 1.2038834951456311e-05,
|
| 100 |
+
"loss": 2.2121,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.13592233009708737,
|
| 105 |
+
"grad_norm": 5.089141368865967,
|
| 106 |
+
"learning_rate": 1.300970873786408e-05,
|
| 107 |
+
"loss": 2.233,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.14563106796116504,
|
| 112 |
+
"grad_norm": 5.0798749923706055,
|
| 113 |
+
"learning_rate": 1.3980582524271846e-05,
|
| 114 |
+
"loss": 2.1688,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.1553398058252427,
|
| 119 |
+
"grad_norm": 5.230806827545166,
|
| 120 |
+
"learning_rate": 1.4951456310679614e-05,
|
| 121 |
+
"loss": 2.2232,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.1650485436893204,
|
| 126 |
+
"grad_norm": 5.30760383605957,
|
| 127 |
+
"learning_rate": 1.592233009708738e-05,
|
| 128 |
+
"loss": 2.1506,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.17475728155339806,
|
| 133 |
+
"grad_norm": 5.008656978607178,
|
| 134 |
+
"learning_rate": 1.6893203883495145e-05,
|
| 135 |
+
"loss": 2.1866,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.18446601941747573,
|
| 140 |
+
"grad_norm": 4.95796537399292,
|
| 141 |
+
"learning_rate": 1.7864077669902916e-05,
|
| 142 |
+
"loss": 2.2156,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.1941747572815534,
|
| 147 |
+
"grad_norm": 4.717769145965576,
|
| 148 |
+
"learning_rate": 1.883495145631068e-05,
|
| 149 |
+
"loss": 2.164,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.20388349514563106,
|
| 154 |
+
"grad_norm": 4.858338832855225,
|
| 155 |
+
"learning_rate": 1.9805825242718447e-05,
|
| 156 |
+
"loss": 2.1291,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.21359223300970873,
|
| 161 |
+
"grad_norm": 5.218167781829834,
|
| 162 |
+
"learning_rate": 1.9913700107874866e-05,
|
| 163 |
+
"loss": 2.1617,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.22330097087378642,
|
| 168 |
+
"grad_norm": 5.097916126251221,
|
| 169 |
+
"learning_rate": 1.9805825242718447e-05,
|
| 170 |
+
"loss": 2.1721,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.23300970873786409,
|
| 175 |
+
"grad_norm": 5.860560417175293,
|
| 176 |
+
"learning_rate": 1.969795037756203e-05,
|
| 177 |
+
"loss": 2.1412,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.24271844660194175,
|
| 182 |
+
"grad_norm": 5.395883560180664,
|
| 183 |
+
"learning_rate": 1.959007551240561e-05,
|
| 184 |
+
"loss": 2.1397,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.2524271844660194,
|
| 189 |
+
"grad_norm": 5.043527126312256,
|
| 190 |
+
"learning_rate": 1.9482200647249193e-05,
|
| 191 |
+
"loss": 2.1314,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.2621359223300971,
|
| 196 |
+
"grad_norm": 4.853712558746338,
|
| 197 |
+
"learning_rate": 1.9374325782092775e-05,
|
| 198 |
+
"loss": 2.118,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.27184466019417475,
|
| 203 |
+
"grad_norm": 5.681634902954102,
|
| 204 |
+
"learning_rate": 1.9266450916936353e-05,
|
| 205 |
+
"loss": 2.1189,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.2815533980582524,
|
| 210 |
+
"grad_norm": 5.401227951049805,
|
| 211 |
+
"learning_rate": 1.9158576051779935e-05,
|
| 212 |
+
"loss": 2.1148,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.2912621359223301,
|
| 217 |
+
"grad_norm": 5.208418369293213,
|
| 218 |
+
"learning_rate": 1.905070118662352e-05,
|
| 219 |
+
"loss": 2.128,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.30097087378640774,
|
| 224 |
+
"grad_norm": 5.307718276977539,
|
| 225 |
+
"learning_rate": 1.89428263214671e-05,
|
| 226 |
+
"loss": 2.0858,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.3106796116504854,
|
| 231 |
+
"grad_norm": 5.279410362243652,
|
| 232 |
+
"learning_rate": 1.883495145631068e-05,
|
| 233 |
+
"loss": 2.1371,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.32038834951456313,
|
| 238 |
+
"grad_norm": 5.151274681091309,
|
| 239 |
+
"learning_rate": 1.8727076591154262e-05,
|
| 240 |
+
"loss": 2.0932,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.3300970873786408,
|
| 245 |
+
"grad_norm": 5.083354473114014,
|
| 246 |
+
"learning_rate": 1.8619201725997844e-05,
|
| 247 |
+
"loss": 2.0623,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.33980582524271846,
|
| 252 |
+
"grad_norm": 5.1322550773620605,
|
| 253 |
+
"learning_rate": 1.8511326860841425e-05,
|
| 254 |
+
"loss": 2.05,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.34951456310679613,
|
| 259 |
+
"grad_norm": 4.970919609069824,
|
| 260 |
+
"learning_rate": 1.8403451995685007e-05,
|
| 261 |
+
"loss": 2.0901,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.3592233009708738,
|
| 266 |
+
"grad_norm": 5.15512752532959,
|
| 267 |
+
"learning_rate": 1.829557713052859e-05,
|
| 268 |
+
"loss": 2.0883,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.36893203883495146,
|
| 273 |
+
"grad_norm": 5.088575839996338,
|
| 274 |
+
"learning_rate": 1.818770226537217e-05,
|
| 275 |
+
"loss": 2.1119,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.3786407766990291,
|
| 280 |
+
"grad_norm": 6.092918872833252,
|
| 281 |
+
"learning_rate": 1.807982740021575e-05,
|
| 282 |
+
"loss": 2.0979,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.3883495145631068,
|
| 287 |
+
"grad_norm": 4.909801483154297,
|
| 288 |
+
"learning_rate": 1.797195253505933e-05,
|
| 289 |
+
"loss": 2.0283,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.39805825242718446,
|
| 294 |
+
"grad_norm": 5.128530025482178,
|
| 295 |
+
"learning_rate": 1.7864077669902916e-05,
|
| 296 |
+
"loss": 2.023,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.4077669902912621,
|
| 301 |
+
"grad_norm": 4.998912811279297,
|
| 302 |
+
"learning_rate": 1.7756202804746498e-05,
|
| 303 |
+
"loss": 2.0207,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.4174757281553398,
|
| 308 |
+
"grad_norm": 5.182358264923096,
|
| 309 |
+
"learning_rate": 1.7648327939590076e-05,
|
| 310 |
+
"loss": 1.9837,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.42718446601941745,
|
| 315 |
+
"grad_norm": 5.3191022872924805,
|
| 316 |
+
"learning_rate": 1.7540453074433658e-05,
|
| 317 |
+
"loss": 2.058,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.4368932038834951,
|
| 322 |
+
"grad_norm": 5.306585788726807,
|
| 323 |
+
"learning_rate": 1.743257820927724e-05,
|
| 324 |
+
"loss": 2.0832,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.44660194174757284,
|
| 329 |
+
"grad_norm": 5.278446197509766,
|
| 330 |
+
"learning_rate": 1.732470334412082e-05,
|
| 331 |
+
"loss": 2.0594,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.4563106796116505,
|
| 336 |
+
"grad_norm": 5.484086990356445,
|
| 337 |
+
"learning_rate": 1.7216828478964403e-05,
|
| 338 |
+
"loss": 2.0763,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.46601941747572817,
|
| 343 |
+
"grad_norm": 5.767387866973877,
|
| 344 |
+
"learning_rate": 1.7108953613807985e-05,
|
| 345 |
+
"loss": 2.0634,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.47572815533980584,
|
| 350 |
+
"grad_norm": 4.96846342086792,
|
| 351 |
+
"learning_rate": 1.7001078748651563e-05,
|
| 352 |
+
"loss": 2.0769,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.4854368932038835,
|
| 357 |
+
"grad_norm": 5.264239311218262,
|
| 358 |
+
"learning_rate": 1.6893203883495145e-05,
|
| 359 |
+
"loss": 2.0604,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.49514563106796117,
|
| 364 |
+
"grad_norm": 5.036663055419922,
|
| 365 |
+
"learning_rate": 1.6785329018338727e-05,
|
| 366 |
+
"loss": 2.1031,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.5048543689320388,
|
| 371 |
+
"grad_norm": 4.875285625457764,
|
| 372 |
+
"learning_rate": 1.6677454153182312e-05,
|
| 373 |
+
"loss": 2.0457,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.5145631067961165,
|
| 378 |
+
"grad_norm": 4.933873653411865,
|
| 379 |
+
"learning_rate": 1.6569579288025894e-05,
|
| 380 |
+
"loss": 2.0312,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.5242718446601942,
|
| 385 |
+
"grad_norm": 5.284345626831055,
|
| 386 |
+
"learning_rate": 1.6461704422869472e-05,
|
| 387 |
+
"loss": 2.0656,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.5339805825242718,
|
| 392 |
+
"grad_norm": 5.3404998779296875,
|
| 393 |
+
"learning_rate": 1.6353829557713054e-05,
|
| 394 |
+
"loss": 2.1049,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.5436893203883495,
|
| 399 |
+
"grad_norm": 5.243639945983887,
|
| 400 |
+
"learning_rate": 1.6245954692556636e-05,
|
| 401 |
+
"loss": 2.0382,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.5533980582524272,
|
| 406 |
+
"grad_norm": 5.110634803771973,
|
| 407 |
+
"learning_rate": 1.6138079827400217e-05,
|
| 408 |
+
"loss": 2.004,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.5631067961165048,
|
| 413 |
+
"grad_norm": 5.063004493713379,
|
| 414 |
+
"learning_rate": 1.60302049622438e-05,
|
| 415 |
+
"loss": 2.0207,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.5728155339805825,
|
| 420 |
+
"grad_norm": 4.7647271156311035,
|
| 421 |
+
"learning_rate": 1.592233009708738e-05,
|
| 422 |
+
"loss": 2.0534,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.5825242718446602,
|
| 427 |
+
"grad_norm": 5.176267147064209,
|
| 428 |
+
"learning_rate": 1.581445523193096e-05,
|
| 429 |
+
"loss": 2.0627,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.5922330097087378,
|
| 434 |
+
"grad_norm": 5.2062225341796875,
|
| 435 |
+
"learning_rate": 1.570658036677454e-05,
|
| 436 |
+
"loss": 2.0063,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.6019417475728155,
|
| 441 |
+
"grad_norm": 5.044838905334473,
|
| 442 |
+
"learning_rate": 1.5598705501618123e-05,
|
| 443 |
+
"loss": 1.9475,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.6116504854368932,
|
| 448 |
+
"grad_norm": 5.029117584228516,
|
| 449 |
+
"learning_rate": 1.5490830636461708e-05,
|
| 450 |
+
"loss": 2.0673,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.6213592233009708,
|
| 455 |
+
"grad_norm": 5.190179347991943,
|
| 456 |
+
"learning_rate": 1.5382955771305286e-05,
|
| 457 |
+
"loss": 2.0176,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.6310679611650486,
|
| 462 |
+
"grad_norm": 4.956365585327148,
|
| 463 |
+
"learning_rate": 1.5275080906148868e-05,
|
| 464 |
+
"loss": 1.9984,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.6407766990291263,
|
| 469 |
+
"grad_norm": 4.972413539886475,
|
| 470 |
+
"learning_rate": 1.516720604099245e-05,
|
| 471 |
+
"loss": 1.9919,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.6504854368932039,
|
| 476 |
+
"grad_norm": 5.319215297698975,
|
| 477 |
+
"learning_rate": 1.5059331175836032e-05,
|
| 478 |
+
"loss": 1.984,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.6601941747572816,
|
| 483 |
+
"grad_norm": 5.120510578155518,
|
| 484 |
+
"learning_rate": 1.4951456310679614e-05,
|
| 485 |
+
"loss": 1.9838,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.6699029126213593,
|
| 490 |
+
"grad_norm": 4.868938446044922,
|
| 491 |
+
"learning_rate": 1.4843581445523194e-05,
|
| 492 |
+
"loss": 2.006,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.6796116504854369,
|
| 497 |
+
"grad_norm": 5.22821044921875,
|
| 498 |
+
"learning_rate": 1.4735706580366775e-05,
|
| 499 |
+
"loss": 2.0099,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.6893203883495146,
|
| 504 |
+
"grad_norm": 4.970730781555176,
|
| 505 |
+
"learning_rate": 1.4627831715210357e-05,
|
| 506 |
+
"loss": 2.0078,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.6990291262135923,
|
| 511 |
+
"grad_norm": 4.913213729858398,
|
| 512 |
+
"learning_rate": 1.4519956850053937e-05,
|
| 513 |
+
"loss": 2.0198,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.7087378640776699,
|
| 518 |
+
"grad_norm": 5.104898452758789,
|
| 519 |
+
"learning_rate": 1.4412081984897519e-05,
|
| 520 |
+
"loss": 2.0224,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.7184466019417476,
|
| 525 |
+
"grad_norm": 4.992263317108154,
|
| 526 |
+
"learning_rate": 1.4304207119741102e-05,
|
| 527 |
+
"loss": 2.0013,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.7281553398058253,
|
| 532 |
+
"grad_norm": 5.0994038581848145,
|
| 533 |
+
"learning_rate": 1.4196332254584684e-05,
|
| 534 |
+
"loss": 1.9542,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.7378640776699029,
|
| 539 |
+
"grad_norm": 5.849913120269775,
|
| 540 |
+
"learning_rate": 1.4088457389428264e-05,
|
| 541 |
+
"loss": 1.9941,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.7475728155339806,
|
| 546 |
+
"grad_norm": 5.09085750579834,
|
| 547 |
+
"learning_rate": 1.3980582524271846e-05,
|
| 548 |
+
"loss": 2.0384,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.7572815533980582,
|
| 553 |
+
"grad_norm": 5.28529167175293,
|
| 554 |
+
"learning_rate": 1.3872707659115428e-05,
|
| 555 |
+
"loss": 2.0251,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.7669902912621359,
|
| 560 |
+
"grad_norm": 5.162165641784668,
|
| 561 |
+
"learning_rate": 1.3764832793959008e-05,
|
| 562 |
+
"loss": 1.9806,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.7766990291262136,
|
| 567 |
+
"grad_norm": 5.865965843200684,
|
| 568 |
+
"learning_rate": 1.365695792880259e-05,
|
| 569 |
+
"loss": 2.0359,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.7864077669902912,
|
| 574 |
+
"grad_norm": 4.936879634857178,
|
| 575 |
+
"learning_rate": 1.3549083063646171e-05,
|
| 576 |
+
"loss": 1.993,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.7961165048543689,
|
| 581 |
+
"grad_norm": 5.331514358520508,
|
| 582 |
+
"learning_rate": 1.3441208198489753e-05,
|
| 583 |
+
"loss": 1.9685,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.8058252427184466,
|
| 588 |
+
"grad_norm": 5.171398639678955,
|
| 589 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 590 |
+
"loss": 1.9885,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.8155339805825242,
|
| 595 |
+
"grad_norm": 4.853579998016357,
|
| 596 |
+
"learning_rate": 1.3225458468176915e-05,
|
| 597 |
+
"loss": 1.9818,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.8252427184466019,
|
| 602 |
+
"grad_norm": 5.196751117706299,
|
| 603 |
+
"learning_rate": 1.3117583603020499e-05,
|
| 604 |
+
"loss": 1.946,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.8349514563106796,
|
| 609 |
+
"grad_norm": 4.931100845336914,
|
| 610 |
+
"learning_rate": 1.300970873786408e-05,
|
| 611 |
+
"loss": 1.9898,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.8446601941747572,
|
| 616 |
+
"grad_norm": 5.232204437255859,
|
| 617 |
+
"learning_rate": 1.290183387270766e-05,
|
| 618 |
+
"loss": 2.0314,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.8543689320388349,
|
| 623 |
+
"grad_norm": 5.175143718719482,
|
| 624 |
+
"learning_rate": 1.2793959007551242e-05,
|
| 625 |
+
"loss": 1.9958,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.8640776699029126,
|
| 630 |
+
"grad_norm": 5.501524925231934,
|
| 631 |
+
"learning_rate": 1.2686084142394824e-05,
|
| 632 |
+
"loss": 1.969,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.8737864077669902,
|
| 637 |
+
"grad_norm": 5.200106620788574,
|
| 638 |
+
"learning_rate": 1.2578209277238404e-05,
|
| 639 |
+
"loss": 1.9624,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.883495145631068,
|
| 644 |
+
"grad_norm": 5.435555934906006,
|
| 645 |
+
"learning_rate": 1.2470334412081986e-05,
|
| 646 |
+
"loss": 2.0057,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.8932038834951457,
|
| 651 |
+
"grad_norm": 5.041926860809326,
|
| 652 |
+
"learning_rate": 1.2362459546925568e-05,
|
| 653 |
+
"loss": 1.9604,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.9029126213592233,
|
| 658 |
+
"grad_norm": 5.86530065536499,
|
| 659 |
+
"learning_rate": 1.2254584681769148e-05,
|
| 660 |
+
"loss": 1.9904,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.912621359223301,
|
| 665 |
+
"grad_norm": 5.039781093597412,
|
| 666 |
+
"learning_rate": 1.214670981661273e-05,
|
| 667 |
+
"loss": 1.9681,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.9223300970873787,
|
| 672 |
+
"grad_norm": 5.195461273193359,
|
| 673 |
+
"learning_rate": 1.2038834951456311e-05,
|
| 674 |
+
"loss": 2.0042,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.9320388349514563,
|
| 679 |
+
"grad_norm": 5.229151725769043,
|
| 680 |
+
"learning_rate": 1.1930960086299891e-05,
|
| 681 |
+
"loss": 2.0039,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.941747572815534,
|
| 686 |
+
"grad_norm": 4.973319053649902,
|
| 687 |
+
"learning_rate": 1.1823085221143475e-05,
|
| 688 |
+
"loss": 2.0065,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.9514563106796117,
|
| 693 |
+
"grad_norm": 5.273900032043457,
|
| 694 |
+
"learning_rate": 1.1715210355987056e-05,
|
| 695 |
+
"loss": 1.9896,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.9611650485436893,
|
| 700 |
+
"grad_norm": 5.269063472747803,
|
| 701 |
+
"learning_rate": 1.1607335490830638e-05,
|
| 702 |
+
"loss": 1.9592,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.970873786407767,
|
| 707 |
+
"grad_norm": 5.100589752197266,
|
| 708 |
+
"learning_rate": 1.149946062567422e-05,
|
| 709 |
+
"loss": 1.9354,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.9805825242718447,
|
| 714 |
+
"grad_norm": 5.232608318328857,
|
| 715 |
+
"learning_rate": 1.1413160733549084e-05,
|
| 716 |
+
"loss": 2.0395,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.9902912621359223,
|
| 721 |
+
"grad_norm": 5.08132791519165,
|
| 722 |
+
"learning_rate": 1.1305285868392666e-05,
|
| 723 |
+
"loss": 1.9437,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 1.0,
|
| 728 |
+
"grad_norm": 5.696261405944824,
|
| 729 |
+
"learning_rate": 1.1197411003236248e-05,
|
| 730 |
+
"loss": 1.9665,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 1.0097087378640777,
|
| 735 |
+
"grad_norm": 5.359734058380127,
|
| 736 |
+
"learning_rate": 1.1089536138079828e-05,
|
| 737 |
+
"loss": 1.841,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 1.0194174757281553,
|
| 742 |
+
"grad_norm": 5.353382587432861,
|
| 743 |
+
"learning_rate": 1.098166127292341e-05,
|
| 744 |
+
"loss": 1.8354,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 1.029126213592233,
|
| 749 |
+
"grad_norm": 5.595281600952148,
|
| 750 |
+
"learning_rate": 1.0873786407766991e-05,
|
| 751 |
+
"loss": 1.8011,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.0388349514563107,
|
| 756 |
+
"grad_norm": 5.603349685668945,
|
| 757 |
+
"learning_rate": 1.0765911542610571e-05,
|
| 758 |
+
"loss": 1.8443,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.0485436893203883,
|
| 763 |
+
"grad_norm": 5.549112319946289,
|
| 764 |
+
"learning_rate": 1.0658036677454153e-05,
|
| 765 |
+
"loss": 1.8022,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 1.058252427184466,
|
| 770 |
+
"grad_norm": 5.524362087249756,
|
| 771 |
+
"learning_rate": 1.0550161812297735e-05,
|
| 772 |
+
"loss": 1.8361,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 1.0679611650485437,
|
| 777 |
+
"grad_norm": 5.287837028503418,
|
| 778 |
+
"learning_rate": 1.0442286947141318e-05,
|
| 779 |
+
"loss": 1.7831,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 1.0776699029126213,
|
| 784 |
+
"grad_norm": 5.503448486328125,
|
| 785 |
+
"learning_rate": 1.03344120819849e-05,
|
| 786 |
+
"loss": 1.816,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 1.087378640776699,
|
| 791 |
+
"grad_norm": 5.447495460510254,
|
| 792 |
+
"learning_rate": 1.022653721682848e-05,
|
| 793 |
+
"loss": 1.7967,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 1.0970873786407767,
|
| 798 |
+
"grad_norm": 5.651370525360107,
|
| 799 |
+
"learning_rate": 1.0118662351672062e-05,
|
| 800 |
+
"loss": 1.767,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 1.1067961165048543,
|
| 805 |
+
"grad_norm": 5.455685138702393,
|
| 806 |
+
"learning_rate": 1.0010787486515644e-05,
|
| 807 |
+
"loss": 1.7986,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 1.116504854368932,
|
| 812 |
+
"grad_norm": 5.773884296417236,
|
| 813 |
+
"learning_rate": 9.902912621359224e-06,
|
| 814 |
+
"loss": 1.8061,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 1.1262135922330097,
|
| 819 |
+
"grad_norm": 5.654228687286377,
|
| 820 |
+
"learning_rate": 9.795037756202806e-06,
|
| 821 |
+
"loss": 1.8051,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 1.1359223300970873,
|
| 826 |
+
"grad_norm": 5.353296756744385,
|
| 827 |
+
"learning_rate": 9.687162891046387e-06,
|
| 828 |
+
"loss": 1.8157,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 1.145631067961165,
|
| 833 |
+
"grad_norm": 5.371259689331055,
|
| 834 |
+
"learning_rate": 9.579288025889967e-06,
|
| 835 |
+
"loss": 1.7957,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 1.1553398058252426,
|
| 840 |
+
"grad_norm": 5.551179885864258,
|
| 841 |
+
"learning_rate": 9.47141316073355e-06,
|
| 842 |
+
"loss": 1.8099,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 1.1650485436893203,
|
| 847 |
+
"grad_norm": 5.763803958892822,
|
| 848 |
+
"learning_rate": 9.363538295577131e-06,
|
| 849 |
+
"loss": 1.8243,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.174757281553398,
|
| 854 |
+
"grad_norm": 5.4469313621521,
|
| 855 |
+
"learning_rate": 9.255663430420713e-06,
|
| 856 |
+
"loss": 1.7813,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.1844660194174756,
|
| 861 |
+
"grad_norm": 5.914862155914307,
|
| 862 |
+
"learning_rate": 9.147788565264294e-06,
|
| 863 |
+
"loss": 1.8308,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.1941747572815533,
|
| 868 |
+
"grad_norm": 5.619472980499268,
|
| 869 |
+
"learning_rate": 9.039913700107874e-06,
|
| 870 |
+
"loss": 1.8566,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.203883495145631,
|
| 875 |
+
"grad_norm": 5.747879981994629,
|
| 876 |
+
"learning_rate": 8.932038834951458e-06,
|
| 877 |
+
"loss": 1.831,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.2135922330097086,
|
| 882 |
+
"grad_norm": 5.515039443969727,
|
| 883 |
+
"learning_rate": 8.824163969795038e-06,
|
| 884 |
+
"loss": 1.8279,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 1.2233009708737863,
|
| 889 |
+
"grad_norm": 5.6780171394348145,
|
| 890 |
+
"learning_rate": 8.71628910463862e-06,
|
| 891 |
+
"loss": 1.7496,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.233009708737864,
|
| 896 |
+
"grad_norm": 5.678586006164551,
|
| 897 |
+
"learning_rate": 8.608414239482202e-06,
|
| 898 |
+
"loss": 1.7862,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.2427184466019416,
|
| 903 |
+
"grad_norm": 5.727756977081299,
|
| 904 |
+
"learning_rate": 8.500539374325782e-06,
|
| 905 |
+
"loss": 1.8364,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.2524271844660193,
|
| 910 |
+
"grad_norm": 5.593883037567139,
|
| 911 |
+
"learning_rate": 8.392664509169363e-06,
|
| 912 |
+
"loss": 1.7775,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.262135922330097,
|
| 917 |
+
"grad_norm": 5.856795310974121,
|
| 918 |
+
"learning_rate": 8.284789644012947e-06,
|
| 919 |
+
"loss": 1.8652,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.2718446601941746,
|
| 924 |
+
"grad_norm": 5.66147518157959,
|
| 925 |
+
"learning_rate": 8.176914778856527e-06,
|
| 926 |
+
"loss": 1.7747,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.2815533980582523,
|
| 931 |
+
"grad_norm": 5.7044291496276855,
|
| 932 |
+
"learning_rate": 8.069039913700109e-06,
|
| 933 |
+
"loss": 1.7781,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.29126213592233,
|
| 938 |
+
"grad_norm": 5.8010149002075195,
|
| 939 |
+
"learning_rate": 7.96116504854369e-06,
|
| 940 |
+
"loss": 1.812,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.3009708737864076,
|
| 945 |
+
"grad_norm": 5.601301670074463,
|
| 946 |
+
"learning_rate": 7.85329018338727e-06,
|
| 947 |
+
"loss": 1.8109,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.3106796116504853,
|
| 952 |
+
"grad_norm": 5.59577751159668,
|
| 953 |
+
"learning_rate": 7.745415318230854e-06,
|
| 954 |
+
"loss": 1.794,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.3203883495145632,
|
| 959 |
+
"grad_norm": 6.064187526702881,
|
| 960 |
+
"learning_rate": 7.637540453074434e-06,
|
| 961 |
+
"loss": 1.7978,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.3300970873786409,
|
| 966 |
+
"grad_norm": 5.477755069732666,
|
| 967 |
+
"learning_rate": 7.529665587918016e-06,
|
| 968 |
+
"loss": 1.7732,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.3398058252427185,
|
| 973 |
+
"grad_norm": 5.672438144683838,
|
| 974 |
+
"learning_rate": 7.421790722761597e-06,
|
| 975 |
+
"loss": 1.8356,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.3495145631067962,
|
| 980 |
+
"grad_norm": 5.968810558319092,
|
| 981 |
+
"learning_rate": 7.3139158576051786e-06,
|
| 982 |
+
"loss": 1.8513,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.3592233009708738,
|
| 987 |
+
"grad_norm": 5.984207630157471,
|
| 988 |
+
"learning_rate": 7.2060409924487595e-06,
|
| 989 |
+
"loss": 1.7821,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.3689320388349515,
|
| 994 |
+
"grad_norm": 6.006514072418213,
|
| 995 |
+
"learning_rate": 7.098166127292342e-06,
|
| 996 |
+
"loss": 1.756,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.3786407766990292,
|
| 1001 |
+
"grad_norm": 5.548986911773682,
|
| 1002 |
+
"learning_rate": 6.990291262135923e-06,
|
| 1003 |
+
"loss": 1.8,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.3883495145631068,
|
| 1008 |
+
"grad_norm": 5.688983917236328,
|
| 1009 |
+
"learning_rate": 6.882416396979504e-06,
|
| 1010 |
+
"loss": 1.7601,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.3980582524271845,
|
| 1015 |
+
"grad_norm": 5.5633225440979,
|
| 1016 |
+
"learning_rate": 6.774541531823086e-06,
|
| 1017 |
+
"loss": 1.7876,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.4077669902912622,
|
| 1022 |
+
"grad_norm": 5.3327226638793945,
|
| 1023 |
+
"learning_rate": 6.666666666666667e-06,
|
| 1024 |
+
"loss": 1.7388,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.4174757281553398,
|
| 1029 |
+
"grad_norm": 5.563536643981934,
|
| 1030 |
+
"learning_rate": 6.558791801510249e-06,
|
| 1031 |
+
"loss": 1.7894,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.4271844660194175,
|
| 1036 |
+
"grad_norm": 5.413880825042725,
|
| 1037 |
+
"learning_rate": 6.45091693635383e-06,
|
| 1038 |
+
"loss": 1.8045,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.4368932038834952,
|
| 1043 |
+
"grad_norm": 5.559710502624512,
|
| 1044 |
+
"learning_rate": 6.343042071197412e-06,
|
| 1045 |
+
"loss": 1.7388,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.4466019417475728,
|
| 1050 |
+
"grad_norm": 5.674643039703369,
|
| 1051 |
+
"learning_rate": 6.235167206040993e-06,
|
| 1052 |
+
"loss": 1.7521,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.4563106796116505,
|
| 1057 |
+
"grad_norm": 6.015341758728027,
|
| 1058 |
+
"learning_rate": 6.127292340884574e-06,
|
| 1059 |
+
"loss": 1.7567,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.4660194174757282,
|
| 1064 |
+
"grad_norm": 5.763010025024414,
|
| 1065 |
+
"learning_rate": 6.0194174757281556e-06,
|
| 1066 |
+
"loss": 1.7494,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.4757281553398058,
|
| 1071 |
+
"grad_norm": 5.727349758148193,
|
| 1072 |
+
"learning_rate": 5.911542610571737e-06,
|
| 1073 |
+
"loss": 1.8048,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.4854368932038835,
|
| 1078 |
+
"grad_norm": 5.473784923553467,
|
| 1079 |
+
"learning_rate": 5.803667745415319e-06,
|
| 1080 |
+
"loss": 1.7469,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.4951456310679612,
|
| 1085 |
+
"grad_norm": 5.847958087921143,
|
| 1086 |
+
"learning_rate": 5.6957928802589e-06,
|
| 1087 |
+
"loss": 1.7803,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.5048543689320388,
|
| 1092 |
+
"grad_norm": 6.08969259262085,
|
| 1093 |
+
"learning_rate": 5.587918015102482e-06,
|
| 1094 |
+
"loss": 1.796,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.5145631067961165,
|
| 1099 |
+
"grad_norm": 5.455092430114746,
|
| 1100 |
+
"learning_rate": 5.480043149946063e-06,
|
| 1101 |
+
"loss": 1.7495,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.5242718446601942,
|
| 1106 |
+
"grad_norm": 5.9275031089782715,
|
| 1107 |
+
"learning_rate": 5.372168284789644e-06,
|
| 1108 |
+
"loss": 1.814,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.5339805825242718,
|
| 1113 |
+
"grad_norm": 5.613204002380371,
|
| 1114 |
+
"learning_rate": 5.264293419633226e-06,
|
| 1115 |
+
"loss": 1.8095,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.5436893203883495,
|
| 1120 |
+
"grad_norm": 5.575292110443115,
|
| 1121 |
+
"learning_rate": 5.156418554476807e-06,
|
| 1122 |
+
"loss": 1.794,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.5533980582524272,
|
| 1127 |
+
"grad_norm": 6.1768107414245605,
|
| 1128 |
+
"learning_rate": 5.048543689320389e-06,
|
| 1129 |
+
"loss": 1.7858,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.5631067961165048,
|
| 1134 |
+
"grad_norm": 5.83579158782959,
|
| 1135 |
+
"learning_rate": 4.94066882416397e-06,
|
| 1136 |
+
"loss": 1.7229,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.5728155339805825,
|
| 1141 |
+
"grad_norm": 5.574371814727783,
|
| 1142 |
+
"learning_rate": 4.832793959007552e-06,
|
| 1143 |
+
"loss": 1.784,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.5825242718446602,
|
| 1148 |
+
"grad_norm": 5.452093124389648,
|
| 1149 |
+
"learning_rate": 4.724919093851133e-06,
|
| 1150 |
+
"loss": 1.7639,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.5922330097087378,
|
| 1155 |
+
"grad_norm": 5.442083358764648,
|
| 1156 |
+
"learning_rate": 4.617044228694714e-06,
|
| 1157 |
+
"loss": 1.8157,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.6019417475728155,
|
| 1162 |
+
"grad_norm": 5.7211079597473145,
|
| 1163 |
+
"learning_rate": 4.509169363538296e-06,
|
| 1164 |
+
"loss": 1.7338,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.6116504854368932,
|
| 1169 |
+
"grad_norm": 6.113297462463379,
|
| 1170 |
+
"learning_rate": 4.401294498381877e-06,
|
| 1171 |
+
"loss": 1.7044,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.6213592233009708,
|
| 1176 |
+
"grad_norm": 5.693146705627441,
|
| 1177 |
+
"learning_rate": 4.293419633225459e-06,
|
| 1178 |
+
"loss": 1.7722,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.6310679611650487,
|
| 1183 |
+
"grad_norm": 5.956842422485352,
|
| 1184 |
+
"learning_rate": 4.1855447680690406e-06,
|
| 1185 |
+
"loss": 1.7618,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.6407766990291264,
|
| 1190 |
+
"grad_norm": 5.646125793457031,
|
| 1191 |
+
"learning_rate": 4.0776699029126215e-06,
|
| 1192 |
+
"loss": 1.7872,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.650485436893204,
|
| 1197 |
+
"grad_norm": 5.913788795471191,
|
| 1198 |
+
"learning_rate": 3.969795037756203e-06,
|
| 1199 |
+
"loss": 1.7913,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.6601941747572817,
|
| 1204 |
+
"grad_norm": 6.05329704284668,
|
| 1205 |
+
"learning_rate": 3.861920172599784e-06,
|
| 1206 |
+
"loss": 1.7771,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.6699029126213594,
|
| 1211 |
+
"grad_norm": 6.139546871185303,
|
| 1212 |
+
"learning_rate": 3.754045307443366e-06,
|
| 1213 |
+
"loss": 1.7966,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.679611650485437,
|
| 1218 |
+
"grad_norm": 6.158768653869629,
|
| 1219 |
+
"learning_rate": 3.6461704422869477e-06,
|
| 1220 |
+
"loss": 1.8097,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.6893203883495147,
|
| 1225 |
+
"grad_norm": 5.726659774780273,
|
| 1226 |
+
"learning_rate": 3.5382955771305286e-06,
|
| 1227 |
+
"loss": 1.7568,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.6990291262135924,
|
| 1232 |
+
"grad_norm": 5.508258819580078,
|
| 1233 |
+
"learning_rate": 3.43042071197411e-06,
|
| 1234 |
+
"loss": 1.8406,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.70873786407767,
|
| 1239 |
+
"grad_norm": 6.076147556304932,
|
| 1240 |
+
"learning_rate": 3.3225458468176918e-06,
|
| 1241 |
+
"loss": 1.7246,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.7184466019417477,
|
| 1246 |
+
"grad_norm": 5.596787452697754,
|
| 1247 |
+
"learning_rate": 3.214670981661273e-06,
|
| 1248 |
+
"loss": 1.8189,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.7281553398058254,
|
| 1253 |
+
"grad_norm": 6.001366138458252,
|
| 1254 |
+
"learning_rate": 3.1067961165048544e-06,
|
| 1255 |
+
"loss": 1.7681,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.737864077669903,
|
| 1260 |
+
"grad_norm": 5.889746189117432,
|
| 1261 |
+
"learning_rate": 2.9989212513484362e-06,
|
| 1262 |
+
"loss": 1.7634,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.7475728155339807,
|
| 1267 |
+
"grad_norm": 5.484528541564941,
|
| 1268 |
+
"learning_rate": 2.8910463861920176e-06,
|
| 1269 |
+
"loss": 1.7928,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.7572815533980584,
|
| 1274 |
+
"grad_norm": 5.366055011749268,
|
| 1275 |
+
"learning_rate": 2.7831715210355993e-06,
|
| 1276 |
+
"loss": 1.7284,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.766990291262136,
|
| 1281 |
+
"grad_norm": 6.179636478424072,
|
| 1282 |
+
"learning_rate": 2.6752966558791803e-06,
|
| 1283 |
+
"loss": 1.7973,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.7766990291262137,
|
| 1288 |
+
"grad_norm": 5.786418437957764,
|
| 1289 |
+
"learning_rate": 2.5674217907227616e-06,
|
| 1290 |
+
"loss": 1.7271,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.7864077669902914,
|
| 1295 |
+
"grad_norm": 5.728253364562988,
|
| 1296 |
+
"learning_rate": 2.4595469255663434e-06,
|
| 1297 |
+
"loss": 1.7947,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.796116504854369,
|
| 1302 |
+
"grad_norm": 5.840207576751709,
|
| 1303 |
+
"learning_rate": 2.3516720604099247e-06,
|
| 1304 |
+
"loss": 1.8411,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.8058252427184467,
|
| 1309 |
+
"grad_norm": 6.026117324829102,
|
| 1310 |
+
"learning_rate": 2.243797195253506e-06,
|
| 1311 |
+
"loss": 1.7801,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.8155339805825244,
|
| 1316 |
+
"grad_norm": 5.574731826782227,
|
| 1317 |
+
"learning_rate": 2.1359223300970874e-06,
|
| 1318 |
+
"loss": 1.8296,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.825242718446602,
|
| 1323 |
+
"grad_norm": 5.741345405578613,
|
| 1324 |
+
"learning_rate": 2.0280474649406688e-06,
|
| 1325 |
+
"loss": 1.8048,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.8349514563106797,
|
| 1330 |
+
"grad_norm": 5.989925384521484,
|
| 1331 |
+
"learning_rate": 1.9201725997842505e-06,
|
| 1332 |
+
"loss": 1.7559,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.8446601941747574,
|
| 1337 |
+
"grad_norm": 5.998227119445801,
|
| 1338 |
+
"learning_rate": 1.812297734627832e-06,
|
| 1339 |
+
"loss": 1.7678,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.854368932038835,
|
| 1344 |
+
"grad_norm": 5.9274420738220215,
|
| 1345 |
+
"learning_rate": 1.7044228694714132e-06,
|
| 1346 |
+
"loss": 1.7886,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.8640776699029127,
|
| 1351 |
+
"grad_norm": 5.719155788421631,
|
| 1352 |
+
"learning_rate": 1.5965480043149948e-06,
|
| 1353 |
+
"loss": 1.7921,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.8737864077669903,
|
| 1358 |
+
"grad_norm": 5.4220147132873535,
|
| 1359 |
+
"learning_rate": 1.4886731391585763e-06,
|
| 1360 |
+
"loss": 1.7813,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.883495145631068,
|
| 1365 |
+
"grad_norm": 5.772354602813721,
|
| 1366 |
+
"learning_rate": 1.3807982740021575e-06,
|
| 1367 |
+
"loss": 1.7606,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.8932038834951457,
|
| 1372 |
+
"grad_norm": 5.864536762237549,
|
| 1373 |
+
"learning_rate": 1.272923408845739e-06,
|
| 1374 |
+
"loss": 1.7354,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.9029126213592233,
|
| 1379 |
+
"grad_norm": 5.453779697418213,
|
| 1380 |
+
"learning_rate": 1.1650485436893206e-06,
|
| 1381 |
+
"loss": 1.7403,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.912621359223301,
|
| 1386 |
+
"grad_norm": 5.6637492179870605,
|
| 1387 |
+
"learning_rate": 1.057173678532902e-06,
|
| 1388 |
+
"loss": 1.7649,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.9223300970873787,
|
| 1393 |
+
"grad_norm": 5.886834621429443,
|
| 1394 |
+
"learning_rate": 9.492988133764834e-07,
|
| 1395 |
+
"loss": 1.8095,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.9320388349514563,
|
| 1400 |
+
"grad_norm": 6.255226135253906,
|
| 1401 |
+
"learning_rate": 8.414239482200648e-07,
|
| 1402 |
+
"loss": 1.7987,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.941747572815534,
|
| 1407 |
+
"grad_norm": 5.694814205169678,
|
| 1408 |
+
"learning_rate": 7.335490830636462e-07,
|
| 1409 |
+
"loss": 1.7715,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.9514563106796117,
|
| 1414 |
+
"grad_norm": 5.4670257568359375,
|
| 1415 |
+
"learning_rate": 6.256742179072277e-07,
|
| 1416 |
+
"loss": 1.737,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.9611650485436893,
|
| 1421 |
+
"grad_norm": 5.776082992553711,
|
| 1422 |
+
"learning_rate": 5.393743257820928e-07,
|
| 1423 |
+
"loss": 1.7518,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.970873786407767,
|
| 1428 |
+
"grad_norm": 5.826039791107178,
|
| 1429 |
+
"learning_rate": 4.314994606256743e-07,
|
| 1430 |
+
"loss": 1.8036,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.9805825242718447,
|
| 1435 |
+
"grad_norm": 5.991348743438721,
|
| 1436 |
+
"learning_rate": 3.2362459546925565e-07,
|
| 1437 |
+
"loss": 1.756,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.9902912621359223,
|
| 1442 |
+
"grad_norm": 5.450629234313965,
|
| 1443 |
+
"learning_rate": 2.1574973031283715e-07,
|
| 1444 |
+
"loss": 1.7953,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 2.0,
|
| 1449 |
+
"grad_norm": 5.462096214294434,
|
| 1450 |
+
"learning_rate": 1.0787486515641857e-07,
|
| 1451 |
+
"loss": 1.7721,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
}
|
| 1454 |
+
],
|
| 1455 |
+
"logging_steps": 10,
|
| 1456 |
+
"max_steps": 2060,
|
| 1457 |
+
"num_input_tokens_seen": 0,
|
| 1458 |
+
"num_train_epochs": 2,
|
| 1459 |
+
"save_steps": 500,
|
| 1460 |
+
"stateful_callbacks": {
|
| 1461 |
+
"TrainerControl": {
|
| 1462 |
+
"args": {
|
| 1463 |
+
"should_epoch_stop": false,
|
| 1464 |
+
"should_evaluate": false,
|
| 1465 |
+
"should_log": false,
|
| 1466 |
+
"should_save": true,
|
| 1467 |
+
"should_training_stop": true
|
| 1468 |
+
},
|
| 1469 |
+
"attributes": {}
|
| 1470 |
+
}
|
| 1471 |
+
},
|
| 1472 |
+
"total_flos": 0.0,
|
| 1473 |
+
"train_batch_size": 32,
|
| 1474 |
+
"trial_name": null,
|
| 1475 |
+
"trial_params": null
|
| 1476 |
+
}
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|