Training checkpoint - Epoch 2, Step 81632
Browse files- checkpoint-81632/1_Pooling/config.json +10 -0
- checkpoint-81632/config.json +25 -0
- checkpoint-81632/config_sentence_transformers.json +14 -0
- checkpoint-81632/modules.json +20 -0
- checkpoint-81632/sentence_bert_config.json +4 -0
- checkpoint-81632/special_tokens_map.json +37 -0
- checkpoint-81632/tokenizer.json +0 -0
- checkpoint-81632/tokenizer_config.json +65 -0
- checkpoint-81632/trainer_state.json +2649 -0
- checkpoint-81632/training_metrics.json +984 -0
- checkpoint-81632/vocab.txt +0 -0
checkpoint-81632/1_Pooling/config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"word_embedding_dimension": 384,
|
| 3 |
+
"pooling_mode_cls_token": false,
|
| 4 |
+
"pooling_mode_mean_tokens": true,
|
| 5 |
+
"pooling_mode_max_tokens": false,
|
| 6 |
+
"pooling_mode_mean_sqrt_len_tokens": false,
|
| 7 |
+
"pooling_mode_weightedmean_tokens": false,
|
| 8 |
+
"pooling_mode_lasttoken": false,
|
| 9 |
+
"include_prompt": true
|
| 10 |
+
}
|
checkpoint-81632/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 |
+
"gradient_checkpointing": false,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 384,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 1536,
|
| 13 |
+
"layer_norm_eps": 1e-12,
|
| 14 |
+
"max_position_embeddings": 512,
|
| 15 |
+
"model_type": "bert",
|
| 16 |
+
"num_attention_heads": 12,
|
| 17 |
+
"num_hidden_layers": 6,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"position_embedding_type": "absolute",
|
| 20 |
+
"torch_dtype": "float32",
|
| 21 |
+
"transformers_version": "4.53.3",
|
| 22 |
+
"type_vocab_size": 2,
|
| 23 |
+
"use_cache": true,
|
| 24 |
+
"vocab_size": 30522
|
| 25 |
+
}
|
checkpoint-81632/config_sentence_transformers.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"__version__": {
|
| 3 |
+
"sentence_transformers": "5.2.0",
|
| 4 |
+
"transformers": "4.53.3",
|
| 5 |
+
"pytorch": "2.6.0+cu124"
|
| 6 |
+
},
|
| 7 |
+
"model_type": "SentenceTransformer",
|
| 8 |
+
"prompts": {
|
| 9 |
+
"query": "",
|
| 10 |
+
"document": ""
|
| 11 |
+
},
|
| 12 |
+
"default_prompt_name": null,
|
| 13 |
+
"similarity_fn_name": "cosine"
|
| 14 |
+
}
|
checkpoint-81632/modules.json
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"idx": 0,
|
| 4 |
+
"name": "0",
|
| 5 |
+
"path": "",
|
| 6 |
+
"type": "sentence_transformers.models.Transformer"
|
| 7 |
+
},
|
| 8 |
+
{
|
| 9 |
+
"idx": 1,
|
| 10 |
+
"name": "1",
|
| 11 |
+
"path": "1_Pooling",
|
| 12 |
+
"type": "sentence_transformers.models.Pooling"
|
| 13 |
+
},
|
| 14 |
+
{
|
| 15 |
+
"idx": 2,
|
| 16 |
+
"name": "2",
|
| 17 |
+
"path": "2_Normalize",
|
| 18 |
+
"type": "sentence_transformers.models.Normalize"
|
| 19 |
+
}
|
| 20 |
+
]
|
checkpoint-81632/sentence_bert_config.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"max_seq_length": 256,
|
| 3 |
+
"do_lower_case": false
|
| 4 |
+
}
|
checkpoint-81632/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 |
+
}
|
checkpoint-81632/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-81632/tokenizer_config.json
ADDED
|
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": false,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_basic_tokenize": true,
|
| 47 |
+
"do_lower_case": true,
|
| 48 |
+
"extra_special_tokens": {},
|
| 49 |
+
"mask_token": "[MASK]",
|
| 50 |
+
"max_length": 128,
|
| 51 |
+
"model_max_length": 256,
|
| 52 |
+
"never_split": null,
|
| 53 |
+
"pad_to_multiple_of": null,
|
| 54 |
+
"pad_token": "[PAD]",
|
| 55 |
+
"pad_token_type_id": 0,
|
| 56 |
+
"padding_side": "right",
|
| 57 |
+
"sep_token": "[SEP]",
|
| 58 |
+
"stride": 0,
|
| 59 |
+
"strip_accents": null,
|
| 60 |
+
"tokenize_chinese_chars": true,
|
| 61 |
+
"tokenizer_class": "BertTokenizer",
|
| 62 |
+
"truncation_side": "right",
|
| 63 |
+
"truncation_strategy": "longest_first",
|
| 64 |
+
"unk_token": "[UNK]"
|
| 65 |
+
}
|
checkpoint-81632/trainer_state.json
ADDED
|
@@ -0,0 +1,2649 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": 81632,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 2.4500196001568013e-05,
|
| 14 |
+
"grad_norm": Infinity,
|
| 15 |
+
"learning_rate": 0.0,
|
| 16 |
+
"loss": 11.8765,
|
| 17 |
+
"step": 1
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.012250098000784006,
|
| 21 |
+
"grad_norm": 52.158992767333984,
|
| 22 |
+
"learning_rate": 3.6457184858507903e-07,
|
| 23 |
+
"loss": 11.14,
|
| 24 |
+
"step": 500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.012250098000784006,
|
| 28 |
+
"eval_cosine_accuracy": 0.9715437889099121,
|
| 29 |
+
"eval_loss": 13.151524543762207,
|
| 30 |
+
"eval_runtime": 48.8891,
|
| 31 |
+
"eval_samples_per_second": 386.753,
|
| 32 |
+
"eval_steps_per_second": 1.514,
|
| 33 |
+
"step": 500
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.02450019600156801,
|
| 37 |
+
"grad_norm": 36.32938003540039,
|
| 38 |
+
"learning_rate": 7.320837927232635e-07,
|
| 39 |
+
"loss": 8.3043,
|
| 40 |
+
"step": 1000
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.02450019600156801,
|
| 44 |
+
"eval_cosine_accuracy": 0.9591735005378723,
|
| 45 |
+
"eval_loss": 8.770788192749023,
|
| 46 |
+
"eval_runtime": 49.7103,
|
| 47 |
+
"eval_samples_per_second": 380.364,
|
| 48 |
+
"eval_steps_per_second": 1.489,
|
| 49 |
+
"step": 1000
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.03675029400235202,
|
| 53 |
+
"grad_norm": 24.141624450683594,
|
| 54 |
+
"learning_rate": 1.099595736861448e-06,
|
| 55 |
+
"loss": 5.25,
|
| 56 |
+
"step": 1500
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.03675029400235202,
|
| 60 |
+
"eval_cosine_accuracy": 0.9491594433784485,
|
| 61 |
+
"eval_loss": 8.737192153930664,
|
| 62 |
+
"eval_runtime": 49.2941,
|
| 63 |
+
"eval_samples_per_second": 383.575,
|
| 64 |
+
"eval_steps_per_second": 1.501,
|
| 65 |
+
"step": 1500
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.04900039200313602,
|
| 69 |
+
"grad_norm": 24.56803321838379,
|
| 70 |
+
"learning_rate": 1.4671076809996327e-06,
|
| 71 |
+
"loss": 4.4706,
|
| 72 |
+
"step": 2000
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.04900039200313602,
|
| 76 |
+
"eval_cosine_accuracy": 0.9433141350746155,
|
| 77 |
+
"eval_loss": 9.47952651977539,
|
| 78 |
+
"eval_runtime": 49.7188,
|
| 79 |
+
"eval_samples_per_second": 380.299,
|
| 80 |
+
"eval_steps_per_second": 1.488,
|
| 81 |
+
"step": 2000
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.06125049000392003,
|
| 85 |
+
"grad_norm": 19.853919982910156,
|
| 86 |
+
"learning_rate": 1.834619625137817e-06,
|
| 87 |
+
"loss": 3.9492,
|
| 88 |
+
"step": 2500
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.06125049000392003,
|
| 92 |
+
"eval_cosine_accuracy": 0.939236044883728,
|
| 93 |
+
"eval_loss": 9.841277122497559,
|
| 94 |
+
"eval_runtime": 48.9761,
|
| 95 |
+
"eval_samples_per_second": 386.066,
|
| 96 |
+
"eval_steps_per_second": 1.511,
|
| 97 |
+
"step": 2500
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.07350058800470403,
|
| 101 |
+
"grad_norm": 18.643808364868164,
|
| 102 |
+
"learning_rate": 2.202131569276001e-06,
|
| 103 |
+
"loss": 3.66,
|
| 104 |
+
"step": 3000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.07350058800470403,
|
| 108 |
+
"eval_cosine_accuracy": 0.9352032542228699,
|
| 109 |
+
"eval_loss": 10.269918441772461,
|
| 110 |
+
"eval_runtime": 49.7068,
|
| 111 |
+
"eval_samples_per_second": 380.391,
|
| 112 |
+
"eval_steps_per_second": 1.489,
|
| 113 |
+
"step": 3000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.08575068600548805,
|
| 117 |
+
"grad_norm": 19.928146362304688,
|
| 118 |
+
"learning_rate": 2.569643513414186e-06,
|
| 119 |
+
"loss": 3.4319,
|
| 120 |
+
"step": 3500
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.08575068600548805,
|
| 124 |
+
"eval_cosine_accuracy": 0.9305813312530518,
|
| 125 |
+
"eval_loss": 10.558481216430664,
|
| 126 |
+
"eval_runtime": 49.1801,
|
| 127 |
+
"eval_samples_per_second": 384.465,
|
| 128 |
+
"eval_steps_per_second": 1.505,
|
| 129 |
+
"step": 3500
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.09800078400627205,
|
| 133 |
+
"grad_norm": 21.710765838623047,
|
| 134 |
+
"learning_rate": 2.9371554575523706e-06,
|
| 135 |
+
"loss": 3.2845,
|
| 136 |
+
"step": 4000
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.09800078400627205,
|
| 140 |
+
"eval_cosine_accuracy": 0.9273641705513,
|
| 141 |
+
"eval_loss": 10.612824440002441,
|
| 142 |
+
"eval_runtime": 49.9816,
|
| 143 |
+
"eval_samples_per_second": 378.299,
|
| 144 |
+
"eval_steps_per_second": 1.481,
|
| 145 |
+
"step": 4000
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.11025088200705606,
|
| 149 |
+
"grad_norm": 12.551642417907715,
|
| 150 |
+
"learning_rate": 3.304667401690555e-06,
|
| 151 |
+
"loss": 3.0736,
|
| 152 |
+
"step": 4500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.11025088200705606,
|
| 156 |
+
"eval_cosine_accuracy": 0.9201141595840454,
|
| 157 |
+
"eval_loss": 11.195393562316895,
|
| 158 |
+
"eval_runtime": 49.2299,
|
| 159 |
+
"eval_samples_per_second": 384.076,
|
| 160 |
+
"eval_steps_per_second": 1.503,
|
| 161 |
+
"step": 4500
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.12250098000784006,
|
| 165 |
+
"grad_norm": 21.684276580810547,
|
| 166 |
+
"learning_rate": 3.6721793458287397e-06,
|
| 167 |
+
"loss": 2.9443,
|
| 168 |
+
"step": 5000
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 0.12250098000784006,
|
| 172 |
+
"eval_cosine_accuracy": 0.9150844812393188,
|
| 173 |
+
"eval_loss": 11.262394905090332,
|
| 174 |
+
"eval_runtime": 49.2542,
|
| 175 |
+
"eval_samples_per_second": 383.886,
|
| 176 |
+
"eval_steps_per_second": 1.502,
|
| 177 |
+
"step": 5000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.13475107800862407,
|
| 181 |
+
"grad_norm": 15.571688652038574,
|
| 182 |
+
"learning_rate": 4.039691289966924e-06,
|
| 183 |
+
"loss": 2.7956,
|
| 184 |
+
"step": 5500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.13475107800862407,
|
| 188 |
+
"eval_cosine_accuracy": 0.9161266684532166,
|
| 189 |
+
"eval_loss": 11.321725845336914,
|
| 190 |
+
"eval_runtime": 50.2817,
|
| 191 |
+
"eval_samples_per_second": 376.041,
|
| 192 |
+
"eval_steps_per_second": 1.472,
|
| 193 |
+
"step": 5500
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.14700117600940807,
|
| 197 |
+
"grad_norm": 23.9124755859375,
|
| 198 |
+
"learning_rate": 4.407203234105109e-06,
|
| 199 |
+
"loss": 2.6543,
|
| 200 |
+
"step": 6000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.14700117600940807,
|
| 204 |
+
"eval_cosine_accuracy": 0.9153110980987549,
|
| 205 |
+
"eval_loss": 11.53966236114502,
|
| 206 |
+
"eval_runtime": 48.8013,
|
| 207 |
+
"eval_samples_per_second": 387.449,
|
| 208 |
+
"eval_steps_per_second": 1.516,
|
| 209 |
+
"step": 6000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.1592512740101921,
|
| 213 |
+
"grad_norm": 18.667434692382812,
|
| 214 |
+
"learning_rate": 4.773980154355016e-06,
|
| 215 |
+
"loss": 2.5272,
|
| 216 |
+
"step": 6500
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.1592512740101921,
|
| 220 |
+
"eval_cosine_accuracy": 0.9076985716819763,
|
| 221 |
+
"eval_loss": 11.740141868591309,
|
| 222 |
+
"eval_runtime": 50.5886,
|
| 223 |
+
"eval_samples_per_second": 373.76,
|
| 224 |
+
"eval_steps_per_second": 1.463,
|
| 225 |
+
"step": 6500
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.1715013720109761,
|
| 229 |
+
"grad_norm": 16.759288787841797,
|
| 230 |
+
"learning_rate": 5.141492098493202e-06,
|
| 231 |
+
"loss": 2.4545,
|
| 232 |
+
"step": 7000
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.1715013720109761,
|
| 236 |
+
"eval_cosine_accuracy": 0.9086048007011414,
|
| 237 |
+
"eval_loss": 11.773887634277344,
|
| 238 |
+
"eval_runtime": 49.7441,
|
| 239 |
+
"eval_samples_per_second": 380.106,
|
| 240 |
+
"eval_steps_per_second": 1.488,
|
| 241 |
+
"step": 7000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.1837514700117601,
|
| 245 |
+
"grad_norm": 11.59648323059082,
|
| 246 |
+
"learning_rate": 5.509004042631386e-06,
|
| 247 |
+
"loss": 2.4065,
|
| 248 |
+
"step": 7500
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.1837514700117601,
|
| 252 |
+
"eval_cosine_accuracy": 0.9028048515319824,
|
| 253 |
+
"eval_loss": 12.162515640258789,
|
| 254 |
+
"eval_runtime": 49.2643,
|
| 255 |
+
"eval_samples_per_second": 383.808,
|
| 256 |
+
"eval_steps_per_second": 1.502,
|
| 257 |
+
"step": 7500
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.1960015680125441,
|
| 261 |
+
"grad_norm": 22.86415672302246,
|
| 262 |
+
"learning_rate": 5.87651598676957e-06,
|
| 263 |
+
"loss": 2.2679,
|
| 264 |
+
"step": 8000
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.1960015680125441,
|
| 268 |
+
"eval_cosine_accuracy": 0.9039376378059387,
|
| 269 |
+
"eval_loss": 11.9808349609375,
|
| 270 |
+
"eval_runtime": 49.5866,
|
| 271 |
+
"eval_samples_per_second": 381.313,
|
| 272 |
+
"eval_steps_per_second": 1.492,
|
| 273 |
+
"step": 8000
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.20825166601332812,
|
| 277 |
+
"grad_norm": 21.962793350219727,
|
| 278 |
+
"learning_rate": 6.243292907019478e-06,
|
| 279 |
+
"loss": 2.1916,
|
| 280 |
+
"step": 8500
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.20825166601332812,
|
| 284 |
+
"eval_cosine_accuracy": 0.9011735916137695,
|
| 285 |
+
"eval_loss": 12.287437438964844,
|
| 286 |
+
"eval_runtime": 49.0894,
|
| 287 |
+
"eval_samples_per_second": 385.175,
|
| 288 |
+
"eval_steps_per_second": 1.507,
|
| 289 |
+
"step": 8500
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.22050176401411212,
|
| 293 |
+
"grad_norm": 19.647785186767578,
|
| 294 |
+
"learning_rate": 6.610804851157663e-06,
|
| 295 |
+
"loss": 2.1577,
|
| 296 |
+
"step": 9000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.22050176401411212,
|
| 300 |
+
"eval_cosine_accuracy": 0.8947392106056213,
|
| 301 |
+
"eval_loss": 12.496896743774414,
|
| 302 |
+
"eval_runtime": 48.9155,
|
| 303 |
+
"eval_samples_per_second": 386.544,
|
| 304 |
+
"eval_steps_per_second": 1.513,
|
| 305 |
+
"step": 9000
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.23275186201489612,
|
| 309 |
+
"grad_norm": 18.5882625579834,
|
| 310 |
+
"learning_rate": 6.978316795295847e-06,
|
| 311 |
+
"loss": 2.0585,
|
| 312 |
+
"step": 9500
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 0.23275186201489612,
|
| 316 |
+
"eval_cosine_accuracy": 0.8911142349243164,
|
| 317 |
+
"eval_loss": 12.531494140625,
|
| 318 |
+
"eval_runtime": 50.0211,
|
| 319 |
+
"eval_samples_per_second": 378.001,
|
| 320 |
+
"eval_steps_per_second": 1.479,
|
| 321 |
+
"step": 9500
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.24500196001568011,
|
| 325 |
+
"grad_norm": 23.30467987060547,
|
| 326 |
+
"learning_rate": 7.345828739434032e-06,
|
| 327 |
+
"loss": 1.9958,
|
| 328 |
+
"step": 10000
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.24500196001568011,
|
| 332 |
+
"eval_cosine_accuracy": 0.8850423693656921,
|
| 333 |
+
"eval_loss": 12.771232604980469,
|
| 334 |
+
"eval_runtime": 49.6132,
|
| 335 |
+
"eval_samples_per_second": 381.109,
|
| 336 |
+
"eval_steps_per_second": 1.492,
|
| 337 |
+
"step": 10000
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 0.2572520580164641,
|
| 341 |
+
"grad_norm": 21.409286499023438,
|
| 342 |
+
"learning_rate": 7.71260565968394e-06,
|
| 343 |
+
"loss": 1.9057,
|
| 344 |
+
"step": 10500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.2572520580164641,
|
| 348 |
+
"eval_cosine_accuracy": 0.8918845653533936,
|
| 349 |
+
"eval_loss": 12.552855491638184,
|
| 350 |
+
"eval_runtime": 48.7739,
|
| 351 |
+
"eval_samples_per_second": 387.667,
|
| 352 |
+
"eval_steps_per_second": 1.517,
|
| 353 |
+
"step": 10500
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.26950215601724814,
|
| 357 |
+
"grad_norm": 21.30998420715332,
|
| 358 |
+
"learning_rate": 8.080117603822123e-06,
|
| 359 |
+
"loss": 1.8862,
|
| 360 |
+
"step": 11000
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.26950215601724814,
|
| 364 |
+
"eval_cosine_accuracy": 0.885540783405304,
|
| 365 |
+
"eval_loss": 12.82436752319336,
|
| 366 |
+
"eval_runtime": 49.3248,
|
| 367 |
+
"eval_samples_per_second": 383.336,
|
| 368 |
+
"eval_steps_per_second": 1.5,
|
| 369 |
+
"step": 11000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.28175225401803217,
|
| 373 |
+
"grad_norm": 19.933103561401367,
|
| 374 |
+
"learning_rate": 8.447629547960307e-06,
|
| 375 |
+
"loss": 1.8489,
|
| 376 |
+
"step": 11500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 0.28175225401803217,
|
| 380 |
+
"eval_cosine_accuracy": 0.8915220499038696,
|
| 381 |
+
"eval_loss": 12.749467849731445,
|
| 382 |
+
"eval_runtime": 48.7596,
|
| 383 |
+
"eval_samples_per_second": 387.78,
|
| 384 |
+
"eval_steps_per_second": 1.518,
|
| 385 |
+
"step": 11500
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 0.29400235201881614,
|
| 389 |
+
"grad_norm": 20.858253479003906,
|
| 390 |
+
"learning_rate": 8.815141492098494e-06,
|
| 391 |
+
"loss": 1.7557,
|
| 392 |
+
"step": 12000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.29400235201881614,
|
| 396 |
+
"eval_cosine_accuracy": 0.8872627019882202,
|
| 397 |
+
"eval_loss": 12.727092742919922,
|
| 398 |
+
"eval_runtime": 49.6118,
|
| 399 |
+
"eval_samples_per_second": 381.119,
|
| 400 |
+
"eval_steps_per_second": 1.492,
|
| 401 |
+
"step": 12000
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.30625245001960016,
|
| 405 |
+
"grad_norm": 16.642915725708008,
|
| 406 |
+
"learning_rate": 9.181918412348402e-06,
|
| 407 |
+
"loss": 1.7018,
|
| 408 |
+
"step": 12500
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.30625245001960016,
|
| 412 |
+
"eval_cosine_accuracy": 0.8861298561096191,
|
| 413 |
+
"eval_loss": 12.634977340698242,
|
| 414 |
+
"eval_runtime": 48.5677,
|
| 415 |
+
"eval_samples_per_second": 389.313,
|
| 416 |
+
"eval_steps_per_second": 1.524,
|
| 417 |
+
"step": 12500
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.3185025480203842,
|
| 421 |
+
"grad_norm": 19.006248474121094,
|
| 422 |
+
"learning_rate": 9.549430356486587e-06,
|
| 423 |
+
"loss": 1.6919,
|
| 424 |
+
"step": 13000
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 0.3185025480203842,
|
| 428 |
+
"eval_cosine_accuracy": 0.8913407921791077,
|
| 429 |
+
"eval_loss": 12.859210014343262,
|
| 430 |
+
"eval_runtime": 49.2157,
|
| 431 |
+
"eval_samples_per_second": 384.187,
|
| 432 |
+
"eval_steps_per_second": 1.504,
|
| 433 |
+
"step": 13000
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.33075264602116816,
|
| 437 |
+
"grad_norm": 17.337366104125977,
|
| 438 |
+
"learning_rate": 9.916942300624771e-06,
|
| 439 |
+
"loss": 1.6186,
|
| 440 |
+
"step": 13500
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.33075264602116816,
|
| 444 |
+
"eval_cosine_accuracy": 0.8802845478057861,
|
| 445 |
+
"eval_loss": 12.690762519836426,
|
| 446 |
+
"eval_runtime": 48.5377,
|
| 447 |
+
"eval_samples_per_second": 389.553,
|
| 448 |
+
"eval_steps_per_second": 1.525,
|
| 449 |
+
"step": 13500
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.3430027440219522,
|
| 453 |
+
"grad_norm": 18.714733123779297,
|
| 454 |
+
"learning_rate": 1.0284454244762956e-05,
|
| 455 |
+
"loss": 1.5835,
|
| 456 |
+
"step": 14000
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 0.3430027440219522,
|
| 460 |
+
"eval_cosine_accuracy": 0.87131267786026,
|
| 461 |
+
"eval_loss": 13.312505722045898,
|
| 462 |
+
"eval_runtime": 48.6142,
|
| 463 |
+
"eval_samples_per_second": 388.94,
|
| 464 |
+
"eval_steps_per_second": 1.522,
|
| 465 |
+
"step": 14000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.35525284202273616,
|
| 469 |
+
"grad_norm": 23.4919376373291,
|
| 470 |
+
"learning_rate": 1.0651231165012862e-05,
|
| 471 |
+
"loss": 1.5391,
|
| 472 |
+
"step": 14500
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.35525284202273616,
|
| 476 |
+
"eval_cosine_accuracy": 0.8672345876693726,
|
| 477 |
+
"eval_loss": 13.471542358398438,
|
| 478 |
+
"eval_runtime": 49.9449,
|
| 479 |
+
"eval_samples_per_second": 378.577,
|
| 480 |
+
"eval_steps_per_second": 1.482,
|
| 481 |
+
"step": 14500
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 0.3675029400235202,
|
| 485 |
+
"grad_norm": 16.173114776611328,
|
| 486 |
+
"learning_rate": 1.1018743109151048e-05,
|
| 487 |
+
"loss": 1.5133,
|
| 488 |
+
"step": 15000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.3675029400235202,
|
| 492 |
+
"eval_cosine_accuracy": 0.8804205060005188,
|
| 493 |
+
"eval_loss": 12.901288032531738,
|
| 494 |
+
"eval_runtime": 49.3453,
|
| 495 |
+
"eval_samples_per_second": 383.177,
|
| 496 |
+
"eval_steps_per_second": 1.5,
|
| 497 |
+
"step": 15000
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 0.3797530380243042,
|
| 501 |
+
"grad_norm": 15.722208976745605,
|
| 502 |
+
"learning_rate": 1.1386255053289233e-05,
|
| 503 |
+
"loss": 1.4875,
|
| 504 |
+
"step": 15500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.3797530380243042,
|
| 508 |
+
"eval_cosine_accuracy": 0.861298680305481,
|
| 509 |
+
"eval_loss": 13.632067680358887,
|
| 510 |
+
"eval_runtime": 48.6675,
|
| 511 |
+
"eval_samples_per_second": 388.514,
|
| 512 |
+
"eval_steps_per_second": 1.521,
|
| 513 |
+
"step": 15500
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.3920031360250882,
|
| 517 |
+
"grad_norm": 20.3407039642334,
|
| 518 |
+
"learning_rate": 1.1753766997427418e-05,
|
| 519 |
+
"loss": 1.3725,
|
| 520 |
+
"step": 16000
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.3920031360250882,
|
| 524 |
+
"eval_cosine_accuracy": 0.876070499420166,
|
| 525 |
+
"eval_loss": 12.429121017456055,
|
| 526 |
+
"eval_runtime": 49.3703,
|
| 527 |
+
"eval_samples_per_second": 382.983,
|
| 528 |
+
"eval_steps_per_second": 1.499,
|
| 529 |
+
"step": 16000
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.4042532340258722,
|
| 533 |
+
"grad_norm": 17.731689453125,
|
| 534 |
+
"learning_rate": 1.2120543917677324e-05,
|
| 535 |
+
"loss": 1.4114,
|
| 536 |
+
"step": 16500
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.4042532340258722,
|
| 540 |
+
"eval_cosine_accuracy": 0.8699533343315125,
|
| 541 |
+
"eval_loss": 13.132347106933594,
|
| 542 |
+
"eval_runtime": 48.6164,
|
| 543 |
+
"eval_samples_per_second": 388.923,
|
| 544 |
+
"eval_steps_per_second": 1.522,
|
| 545 |
+
"step": 16500
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 0.41650333202665624,
|
| 549 |
+
"grad_norm": 11.438084602355957,
|
| 550 |
+
"learning_rate": 1.2488055861815509e-05,
|
| 551 |
+
"loss": 1.4118,
|
| 552 |
+
"step": 17000
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 0.41650333202665624,
|
| 556 |
+
"eval_cosine_accuracy": 0.8669173717498779,
|
| 557 |
+
"eval_loss": 13.065674781799316,
|
| 558 |
+
"eval_runtime": 48.6001,
|
| 559 |
+
"eval_samples_per_second": 389.053,
|
| 560 |
+
"eval_steps_per_second": 1.523,
|
| 561 |
+
"step": 17000
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 0.4287534300274402,
|
| 565 |
+
"grad_norm": 15.414648056030273,
|
| 566 |
+
"learning_rate": 1.2855567805953693e-05,
|
| 567 |
+
"loss": 1.3891,
|
| 568 |
+
"step": 17500
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 0.4287534300274402,
|
| 572 |
+
"eval_cosine_accuracy": 0.8608002066612244,
|
| 573 |
+
"eval_loss": 13.467134475708008,
|
| 574 |
+
"eval_runtime": 50.2611,
|
| 575 |
+
"eval_samples_per_second": 376.196,
|
| 576 |
+
"eval_steps_per_second": 1.472,
|
| 577 |
+
"step": 17500
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.44100352802822423,
|
| 581 |
+
"grad_norm": 9.465897560119629,
|
| 582 |
+
"learning_rate": 1.3223079750091878e-05,
|
| 583 |
+
"loss": 1.3271,
|
| 584 |
+
"step": 18000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.44100352802822423,
|
| 588 |
+
"eval_cosine_accuracy": 0.8644705414772034,
|
| 589 |
+
"eval_loss": 13.483769416809082,
|
| 590 |
+
"eval_runtime": 49.6713,
|
| 591 |
+
"eval_samples_per_second": 380.662,
|
| 592 |
+
"eval_steps_per_second": 1.49,
|
| 593 |
+
"step": 18000
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 0.4532536260290082,
|
| 597 |
+
"grad_norm": 11.63443374633789,
|
| 598 |
+
"learning_rate": 1.3589856670341788e-05,
|
| 599 |
+
"loss": 1.3291,
|
| 600 |
+
"step": 18500
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 0.4532536260290082,
|
| 604 |
+
"eval_cosine_accuracy": 0.8525533676147461,
|
| 605 |
+
"eval_loss": 13.853730201721191,
|
| 606 |
+
"eval_runtime": 48.8617,
|
| 607 |
+
"eval_samples_per_second": 386.97,
|
| 608 |
+
"eval_steps_per_second": 1.514,
|
| 609 |
+
"step": 18500
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.46550372402979223,
|
| 613 |
+
"grad_norm": 16.112844467163086,
|
| 614 |
+
"learning_rate": 1.3957368614479972e-05,
|
| 615 |
+
"loss": 1.3198,
|
| 616 |
+
"step": 19000
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 0.46550372402979223,
|
| 620 |
+
"eval_cosine_accuracy": 0.8601658344268799,
|
| 621 |
+
"eval_loss": 13.518957138061523,
|
| 622 |
+
"eval_runtime": 49.367,
|
| 623 |
+
"eval_samples_per_second": 383.009,
|
| 624 |
+
"eval_steps_per_second": 1.499,
|
| 625 |
+
"step": 19000
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.47775382203057626,
|
| 629 |
+
"grad_norm": 14.382566452026367,
|
| 630 |
+
"learning_rate": 1.4324880558618155e-05,
|
| 631 |
+
"loss": 1.2659,
|
| 632 |
+
"step": 19500
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.47775382203057626,
|
| 636 |
+
"eval_cosine_accuracy": 0.8497439622879028,
|
| 637 |
+
"eval_loss": 14.119278907775879,
|
| 638 |
+
"eval_runtime": 48.6599,
|
| 639 |
+
"eval_samples_per_second": 388.574,
|
| 640 |
+
"eval_steps_per_second": 1.521,
|
| 641 |
+
"step": 19500
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.49000392003136023,
|
| 645 |
+
"grad_norm": 16.34287452697754,
|
| 646 |
+
"learning_rate": 1.469239250275634e-05,
|
| 647 |
+
"loss": 1.2598,
|
| 648 |
+
"step": 20000
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 0.49000392003136023,
|
| 652 |
+
"eval_cosine_accuracy": 0.8596221208572388,
|
| 653 |
+
"eval_loss": 13.421476364135742,
|
| 654 |
+
"eval_runtime": 49.3372,
|
| 655 |
+
"eval_samples_per_second": 383.241,
|
| 656 |
+
"eval_steps_per_second": 1.5,
|
| 657 |
+
"step": 20000
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 0.5022540180321443,
|
| 661 |
+
"grad_norm": 14.337957382202148,
|
| 662 |
+
"learning_rate": 1.5059169423006248e-05,
|
| 663 |
+
"loss": 1.223,
|
| 664 |
+
"step": 20500
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 0.5022540180321443,
|
| 668 |
+
"eval_cosine_accuracy": 0.8446236848831177,
|
| 669 |
+
"eval_loss": 13.997146606445312,
|
| 670 |
+
"eval_runtime": 48.7599,
|
| 671 |
+
"eval_samples_per_second": 387.778,
|
| 672 |
+
"eval_steps_per_second": 1.518,
|
| 673 |
+
"step": 20500
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 0.5145041160329282,
|
| 677 |
+
"grad_norm": 25.929706573486328,
|
| 678 |
+
"learning_rate": 1.5426681367144433e-05,
|
| 679 |
+
"loss": 1.2106,
|
| 680 |
+
"step": 21000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.5145041160329282,
|
| 684 |
+
"eval_cosine_accuracy": 0.863473653793335,
|
| 685 |
+
"eval_loss": 13.374435424804688,
|
| 686 |
+
"eval_runtime": 48.8876,
|
| 687 |
+
"eval_samples_per_second": 386.765,
|
| 688 |
+
"eval_steps_per_second": 1.514,
|
| 689 |
+
"step": 21000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.5267542140337123,
|
| 693 |
+
"grad_norm": 17.48406219482422,
|
| 694 |
+
"learning_rate": 1.579419331128262e-05,
|
| 695 |
+
"loss": 1.1485,
|
| 696 |
+
"step": 21500
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.5267542140337123,
|
| 700 |
+
"eval_cosine_accuracy": 0.857628345489502,
|
| 701 |
+
"eval_loss": 13.285289764404297,
|
| 702 |
+
"eval_runtime": 50.3122,
|
| 703 |
+
"eval_samples_per_second": 375.814,
|
| 704 |
+
"eval_steps_per_second": 1.471,
|
| 705 |
+
"step": 21500
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 0.5390043120344963,
|
| 709 |
+
"grad_norm": 5.116608619689941,
|
| 710 |
+
"learning_rate": 1.61617052554208e-05,
|
| 711 |
+
"loss": 1.109,
|
| 712 |
+
"step": 22000
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 0.5390043120344963,
|
| 716 |
+
"eval_cosine_accuracy": 0.8546377420425415,
|
| 717 |
+
"eval_loss": 13.88798713684082,
|
| 718 |
+
"eval_runtime": 49.3125,
|
| 719 |
+
"eval_samples_per_second": 383.433,
|
| 720 |
+
"eval_steps_per_second": 1.501,
|
| 721 |
+
"step": 22000
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 0.5512544100352803,
|
| 725 |
+
"grad_norm": 0.0,
|
| 726 |
+
"learning_rate": 1.652848217567071e-05,
|
| 727 |
+
"loss": 1.0985,
|
| 728 |
+
"step": 22500
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 0.5512544100352803,
|
| 732 |
+
"eval_cosine_accuracy": 0.8662376999855042,
|
| 733 |
+
"eval_loss": 13.286218643188477,
|
| 734 |
+
"eval_runtime": 49.0486,
|
| 735 |
+
"eval_samples_per_second": 385.496,
|
| 736 |
+
"eval_steps_per_second": 1.509,
|
| 737 |
+
"step": 22500
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.5635045080360643,
|
| 741 |
+
"grad_norm": 13.54675006866455,
|
| 742 |
+
"learning_rate": 1.6895994119808893e-05,
|
| 743 |
+
"loss": 1.1176,
|
| 744 |
+
"step": 23000
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.5635045080360643,
|
| 748 |
+
"eval_cosine_accuracy": 0.8439440131187439,
|
| 749 |
+
"eval_loss": 14.143769264221191,
|
| 750 |
+
"eval_runtime": 49.55,
|
| 751 |
+
"eval_samples_per_second": 381.594,
|
| 752 |
+
"eval_steps_per_second": 1.493,
|
| 753 |
+
"step": 23000
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.5757546060368482,
|
| 757 |
+
"grad_norm": 14.270753860473633,
|
| 758 |
+
"learning_rate": 1.726350606394708e-05,
|
| 759 |
+
"loss": 1.1057,
|
| 760 |
+
"step": 23500
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 0.5757546060368482,
|
| 764 |
+
"eval_cosine_accuracy": 0.8401830792427063,
|
| 765 |
+
"eval_loss": 14.088532447814941,
|
| 766 |
+
"eval_runtime": 48.924,
|
| 767 |
+
"eval_samples_per_second": 386.477,
|
| 768 |
+
"eval_steps_per_second": 1.513,
|
| 769 |
+
"step": 23500
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 0.5880047040376323,
|
| 773 |
+
"grad_norm": 10.868962287902832,
|
| 774 |
+
"learning_rate": 1.7630282984196987e-05,
|
| 775 |
+
"loss": 1.0746,
|
| 776 |
+
"step": 24000
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 0.5880047040376323,
|
| 780 |
+
"eval_cosine_accuracy": 0.848339319229126,
|
| 781 |
+
"eval_loss": 13.797255516052246,
|
| 782 |
+
"eval_runtime": 49.6585,
|
| 783 |
+
"eval_samples_per_second": 380.761,
|
| 784 |
+
"eval_steps_per_second": 1.49,
|
| 785 |
+
"step": 24000
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 0.6002548020384163,
|
| 789 |
+
"grad_norm": 9.627524375915527,
|
| 790 |
+
"learning_rate": 1.799779492833517e-05,
|
| 791 |
+
"loss": 1.1169,
|
| 792 |
+
"step": 24500
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 0.6002548020384163,
|
| 796 |
+
"eval_cosine_accuracy": 0.8492002487182617,
|
| 797 |
+
"eval_loss": 13.715460777282715,
|
| 798 |
+
"eval_runtime": 49.1515,
|
| 799 |
+
"eval_samples_per_second": 384.688,
|
| 800 |
+
"eval_steps_per_second": 1.506,
|
| 801 |
+
"step": 24500
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.6125049000392003,
|
| 805 |
+
"grad_norm": 11.183845520019531,
|
| 806 |
+
"learning_rate": 1.8365306872473356e-05,
|
| 807 |
+
"loss": 1.0609,
|
| 808 |
+
"step": 25000
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.6125049000392003,
|
| 812 |
+
"eval_cosine_accuracy": 0.8482486605644226,
|
| 813 |
+
"eval_loss": 14.3268461227417,
|
| 814 |
+
"eval_runtime": 48.8836,
|
| 815 |
+
"eval_samples_per_second": 386.797,
|
| 816 |
+
"eval_steps_per_second": 1.514,
|
| 817 |
+
"step": 25000
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 0.6247549980399844,
|
| 821 |
+
"grad_norm": 13.244708061218262,
|
| 822 |
+
"learning_rate": 1.873281881661154e-05,
|
| 823 |
+
"loss": 1.0227,
|
| 824 |
+
"step": 25500
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 0.6247549980399844,
|
| 828 |
+
"eval_cosine_accuracy": 0.8472517728805542,
|
| 829 |
+
"eval_loss": 13.635766983032227,
|
| 830 |
+
"eval_runtime": 50.3809,
|
| 831 |
+
"eval_samples_per_second": 375.301,
|
| 832 |
+
"eval_steps_per_second": 1.469,
|
| 833 |
+
"step": 25500
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 0.6370050960407684,
|
| 837 |
+
"grad_norm": 9.301331520080566,
|
| 838 |
+
"learning_rate": 1.9100330760749725e-05,
|
| 839 |
+
"loss": 1.0461,
|
| 840 |
+
"step": 26000
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 0.6370050960407684,
|
| 844 |
+
"eval_cosine_accuracy": 0.835198700428009,
|
| 845 |
+
"eval_loss": 14.338939666748047,
|
| 846 |
+
"eval_runtime": 49.6306,
|
| 847 |
+
"eval_samples_per_second": 380.974,
|
| 848 |
+
"eval_steps_per_second": 1.491,
|
| 849 |
+
"step": 26000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.6492551940415523,
|
| 853 |
+
"grad_norm": 15.854827880859375,
|
| 854 |
+
"learning_rate": 1.946784270488791e-05,
|
| 855 |
+
"loss": 0.9854,
|
| 856 |
+
"step": 26500
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.6492551940415523,
|
| 860 |
+
"eval_cosine_accuracy": 0.8212424516677856,
|
| 861 |
+
"eval_loss": 14.60715103149414,
|
| 862 |
+
"eval_runtime": 48.772,
|
| 863 |
+
"eval_samples_per_second": 387.682,
|
| 864 |
+
"eval_steps_per_second": 1.517,
|
| 865 |
+
"step": 26500
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 0.6615052920423363,
|
| 869 |
+
"grad_norm": 11.324312210083008,
|
| 870 |
+
"learning_rate": 1.9835354649026095e-05,
|
| 871 |
+
"loss": 1.0381,
|
| 872 |
+
"step": 27000
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 0.6615052920423363,
|
| 876 |
+
"eval_cosine_accuracy": 0.8262268304824829,
|
| 877 |
+
"eval_loss": 14.688072204589844,
|
| 878 |
+
"eval_runtime": 48.851,
|
| 879 |
+
"eval_samples_per_second": 387.054,
|
| 880 |
+
"eval_steps_per_second": 1.515,
|
| 881 |
+
"step": 27000
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 0.6737553900431204,
|
| 885 |
+
"grad_norm": 10.2542142868042,
|
| 886 |
+
"learning_rate": 2.0202866593164277e-05,
|
| 887 |
+
"loss": 0.9547,
|
| 888 |
+
"step": 27500
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 0.6737553900431204,
|
| 892 |
+
"eval_cosine_accuracy": 0.8329330682754517,
|
| 893 |
+
"eval_loss": 14.224186897277832,
|
| 894 |
+
"eval_runtime": 50.0735,
|
| 895 |
+
"eval_samples_per_second": 377.605,
|
| 896 |
+
"eval_steps_per_second": 1.478,
|
| 897 |
+
"step": 27500
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 0.6860054880439044,
|
| 901 |
+
"grad_norm": 10.368769645690918,
|
| 902 |
+
"learning_rate": 2.0569643513414186e-05,
|
| 903 |
+
"loss": 0.9755,
|
| 904 |
+
"step": 28000
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 0.6860054880439044,
|
| 908 |
+
"eval_cosine_accuracy": 0.8453940153121948,
|
| 909 |
+
"eval_loss": 14.43342399597168,
|
| 910 |
+
"eval_runtime": 51.4882,
|
| 911 |
+
"eval_samples_per_second": 367.229,
|
| 912 |
+
"eval_steps_per_second": 1.437,
|
| 913 |
+
"step": 28000
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.6982555860446884,
|
| 917 |
+
"grad_norm": 5.451234340667725,
|
| 918 |
+
"learning_rate": 2.0936420433664094e-05,
|
| 919 |
+
"loss": 0.9263,
|
| 920 |
+
"step": 28500
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.6982555860446884,
|
| 924 |
+
"eval_cosine_accuracy": 0.8401377201080322,
|
| 925 |
+
"eval_loss": 13.982333183288574,
|
| 926 |
+
"eval_runtime": 48.8777,
|
| 927 |
+
"eval_samples_per_second": 386.843,
|
| 928 |
+
"eval_steps_per_second": 1.514,
|
| 929 |
+
"step": 28500
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 0.7105056840454723,
|
| 933 |
+
"grad_norm": 7.560960292816162,
|
| 934 |
+
"learning_rate": 2.1303932377802277e-05,
|
| 935 |
+
"loss": 0.9208,
|
| 936 |
+
"step": 29000
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 0.7105056840454723,
|
| 940 |
+
"eval_cosine_accuracy": 0.8279939889907837,
|
| 941 |
+
"eval_loss": 14.23220443725586,
|
| 942 |
+
"eval_runtime": 48.9779,
|
| 943 |
+
"eval_samples_per_second": 386.052,
|
| 944 |
+
"eval_steps_per_second": 1.511,
|
| 945 |
+
"step": 29000
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 0.7227557820462563,
|
| 949 |
+
"grad_norm": 0.0,
|
| 950 |
+
"learning_rate": 2.1671444321940463e-05,
|
| 951 |
+
"loss": 0.9248,
|
| 952 |
+
"step": 29500
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 0.7227557820462563,
|
| 956 |
+
"eval_cosine_accuracy": 0.8259549736976624,
|
| 957 |
+
"eval_loss": 14.843314170837402,
|
| 958 |
+
"eval_runtime": 50.5202,
|
| 959 |
+
"eval_samples_per_second": 374.266,
|
| 960 |
+
"eval_steps_per_second": 1.465,
|
| 961 |
+
"step": 29500
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.7350058800470404,
|
| 965 |
+
"grad_norm": 13.433263778686523,
|
| 966 |
+
"learning_rate": 2.2038956266078646e-05,
|
| 967 |
+
"loss": 0.9535,
|
| 968 |
+
"step": 30000
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.7350058800470404,
|
| 972 |
+
"eval_cosine_accuracy": 0.8178893327713013,
|
| 973 |
+
"eval_loss": 14.670516014099121,
|
| 974 |
+
"eval_runtime": 48.7315,
|
| 975 |
+
"eval_samples_per_second": 388.003,
|
| 976 |
+
"eval_steps_per_second": 1.519,
|
| 977 |
+
"step": 30000
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 0.7472559780478244,
|
| 981 |
+
"grad_norm": 0.0,
|
| 982 |
+
"learning_rate": 2.2406468210216832e-05,
|
| 983 |
+
"loss": 0.9203,
|
| 984 |
+
"step": 30500
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 0.7472559780478244,
|
| 988 |
+
"eval_cosine_accuracy": 0.813086211681366,
|
| 989 |
+
"eval_loss": 14.975008964538574,
|
| 990 |
+
"eval_runtime": 48.8999,
|
| 991 |
+
"eval_samples_per_second": 386.667,
|
| 992 |
+
"eval_steps_per_second": 1.513,
|
| 993 |
+
"step": 30500
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 0.7595060760486084,
|
| 997 |
+
"grad_norm": 5.927606582641602,
|
| 998 |
+
"learning_rate": 2.2773980154355015e-05,
|
| 999 |
+
"loss": 0.8704,
|
| 1000 |
+
"step": 31000
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 0.7595060760486084,
|
| 1004 |
+
"eval_cosine_accuracy": 0.8202909231185913,
|
| 1005 |
+
"eval_loss": 14.855353355407715,
|
| 1006 |
+
"eval_runtime": 49.6626,
|
| 1007 |
+
"eval_samples_per_second": 380.729,
|
| 1008 |
+
"eval_steps_per_second": 1.49,
|
| 1009 |
+
"step": 31000
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 0.7717561740493923,
|
| 1013 |
+
"grad_norm": 0.0,
|
| 1014 |
+
"learning_rate": 2.31414920984932e-05,
|
| 1015 |
+
"loss": 0.8854,
|
| 1016 |
+
"step": 31500
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 0.7717561740493923,
|
| 1020 |
+
"eval_cosine_accuracy": 0.8221487402915955,
|
| 1021 |
+
"eval_loss": 14.634927749633789,
|
| 1022 |
+
"eval_runtime": 48.9593,
|
| 1023 |
+
"eval_samples_per_second": 386.199,
|
| 1024 |
+
"eval_steps_per_second": 1.511,
|
| 1025 |
+
"step": 31500
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.7840062720501764,
|
| 1029 |
+
"grad_norm": 8.689051628112793,
|
| 1030 |
+
"learning_rate": 2.3509004042631384e-05,
|
| 1031 |
+
"loss": 0.8855,
|
| 1032 |
+
"step": 32000
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.7840062720501764,
|
| 1036 |
+
"eval_cosine_accuracy": 0.8112737536430359,
|
| 1037 |
+
"eval_loss": 15.120546340942383,
|
| 1038 |
+
"eval_runtime": 48.7938,
|
| 1039 |
+
"eval_samples_per_second": 387.508,
|
| 1040 |
+
"eval_steps_per_second": 1.517,
|
| 1041 |
+
"step": 32000
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 0.7962563700509604,
|
| 1045 |
+
"grad_norm": 0.0,
|
| 1046 |
+
"learning_rate": 2.387651598676957e-05,
|
| 1047 |
+
"loss": 0.8584,
|
| 1048 |
+
"step": 32500
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 0.7962563700509604,
|
| 1052 |
+
"eval_cosine_accuracy": 0.8139471411705017,
|
| 1053 |
+
"eval_loss": 14.621186256408691,
|
| 1054 |
+
"eval_runtime": 49.3901,
|
| 1055 |
+
"eval_samples_per_second": 382.83,
|
| 1056 |
+
"eval_steps_per_second": 1.498,
|
| 1057 |
+
"step": 32500
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 0.8085064680517444,
|
| 1061 |
+
"grad_norm": 8.683895111083984,
|
| 1062 |
+
"learning_rate": 2.4244027930907757e-05,
|
| 1063 |
+
"loss": 0.8361,
|
| 1064 |
+
"step": 33000
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 0.8085064680517444,
|
| 1068 |
+
"eval_cosine_accuracy": 0.8144456148147583,
|
| 1069 |
+
"eval_loss": 14.621684074401855,
|
| 1070 |
+
"eval_runtime": 48.7922,
|
| 1071 |
+
"eval_samples_per_second": 387.521,
|
| 1072 |
+
"eval_steps_per_second": 1.517,
|
| 1073 |
+
"step": 33000
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.8207565660525284,
|
| 1077 |
+
"grad_norm": 3.0560576915740967,
|
| 1078 |
+
"learning_rate": 2.461080485115766e-05,
|
| 1079 |
+
"loss": 0.8283,
|
| 1080 |
+
"step": 33500
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.8207565660525284,
|
| 1084 |
+
"eval_cosine_accuracy": 0.8181612491607666,
|
| 1085 |
+
"eval_loss": 15.029080390930176,
|
| 1086 |
+
"eval_runtime": 49.966,
|
| 1087 |
+
"eval_samples_per_second": 378.418,
|
| 1088 |
+
"eval_steps_per_second": 1.481,
|
| 1089 |
+
"step": 33500
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 0.8330066640533125,
|
| 1093 |
+
"grad_norm": 11.082379341125488,
|
| 1094 |
+
"learning_rate": 2.4978316795295848e-05,
|
| 1095 |
+
"loss": 0.8749,
|
| 1096 |
+
"step": 34000
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 0.8330066640533125,
|
| 1100 |
+
"eval_cosine_accuracy": 0.8164846897125244,
|
| 1101 |
+
"eval_loss": 14.72106647491455,
|
| 1102 |
+
"eval_runtime": 49.4031,
|
| 1103 |
+
"eval_samples_per_second": 382.729,
|
| 1104 |
+
"eval_steps_per_second": 1.498,
|
| 1105 |
+
"step": 34000
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 0.8452567620540964,
|
| 1109 |
+
"grad_norm": 4.894160270690918,
|
| 1110 |
+
"learning_rate": 2.534582873943403e-05,
|
| 1111 |
+
"loss": 0.8387,
|
| 1112 |
+
"step": 34500
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 0.8452567620540964,
|
| 1116 |
+
"eval_cosine_accuracy": 0.8250033855438232,
|
| 1117 |
+
"eval_loss": 14.777338981628418,
|
| 1118 |
+
"eval_runtime": 48.8507,
|
| 1119 |
+
"eval_samples_per_second": 387.057,
|
| 1120 |
+
"eval_steps_per_second": 1.515,
|
| 1121 |
+
"step": 34500
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 0.8575068600548804,
|
| 1125 |
+
"grad_norm": 9.33103084564209,
|
| 1126 |
+
"learning_rate": 2.5713340683572217e-05,
|
| 1127 |
+
"loss": 0.8256,
|
| 1128 |
+
"step": 35000
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 0.8575068600548804,
|
| 1132 |
+
"eval_cosine_accuracy": 0.8253206014633179,
|
| 1133 |
+
"eval_loss": 13.815231323242188,
|
| 1134 |
+
"eval_runtime": 49.4383,
|
| 1135 |
+
"eval_samples_per_second": 382.457,
|
| 1136 |
+
"eval_steps_per_second": 1.497,
|
| 1137 |
+
"step": 35000
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.8697569580556644,
|
| 1141 |
+
"grad_norm": 9.813937187194824,
|
| 1142 |
+
"learning_rate": 2.6080117603822125e-05,
|
| 1143 |
+
"loss": 0.7807,
|
| 1144 |
+
"step": 35500
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.8697569580556644,
|
| 1148 |
+
"eval_cosine_accuracy": 0.8259549736976624,
|
| 1149 |
+
"eval_loss": 14.613081932067871,
|
| 1150 |
+
"eval_runtime": 48.7095,
|
| 1151 |
+
"eval_samples_per_second": 388.179,
|
| 1152 |
+
"eval_steps_per_second": 1.519,
|
| 1153 |
+
"step": 35500
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 0.8820070560564485,
|
| 1157 |
+
"grad_norm": 11.17261028289795,
|
| 1158 |
+
"learning_rate": 2.6447629547960308e-05,
|
| 1159 |
+
"loss": 0.8049,
|
| 1160 |
+
"step": 36000
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 0.8820070560564485,
|
| 1164 |
+
"eval_cosine_accuracy": 0.8182971477508545,
|
| 1165 |
+
"eval_loss": 14.53075885772705,
|
| 1166 |
+
"eval_runtime": 48.8791,
|
| 1167 |
+
"eval_samples_per_second": 386.832,
|
| 1168 |
+
"eval_steps_per_second": 1.514,
|
| 1169 |
+
"step": 36000
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 0.8942571540572325,
|
| 1173 |
+
"grad_norm": 12.27320384979248,
|
| 1174 |
+
"learning_rate": 2.6815141492098494e-05,
|
| 1175 |
+
"loss": 0.7933,
|
| 1176 |
+
"step": 36500
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 0.8942571540572325,
|
| 1180 |
+
"eval_cosine_accuracy": 0.8179346323013306,
|
| 1181 |
+
"eval_loss": 14.517987251281738,
|
| 1182 |
+
"eval_runtime": 49.9587,
|
| 1183 |
+
"eval_samples_per_second": 378.473,
|
| 1184 |
+
"eval_steps_per_second": 1.481,
|
| 1185 |
+
"step": 36500
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.9065072520580164,
|
| 1189 |
+
"grad_norm": 6.898131370544434,
|
| 1190 |
+
"learning_rate": 2.7182653436236677e-05,
|
| 1191 |
+
"loss": 0.7805,
|
| 1192 |
+
"step": 37000
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.9065072520580164,
|
| 1196 |
+
"eval_cosine_accuracy": 0.8172096610069275,
|
| 1197 |
+
"eval_loss": 14.480570793151855,
|
| 1198 |
+
"eval_runtime": 49.2849,
|
| 1199 |
+
"eval_samples_per_second": 383.647,
|
| 1200 |
+
"eval_steps_per_second": 1.501,
|
| 1201 |
+
"step": 37000
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 0.9187573500588004,
|
| 1205 |
+
"grad_norm": 14.374473571777344,
|
| 1206 |
+
"learning_rate": 2.754943035648659e-05,
|
| 1207 |
+
"loss": 0.7737,
|
| 1208 |
+
"step": 37500
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 0.9187573500588004,
|
| 1212 |
+
"eval_cosine_accuracy": 0.8054737448692322,
|
| 1213 |
+
"eval_loss": 15.179258346557617,
|
| 1214 |
+
"eval_runtime": 48.8052,
|
| 1215 |
+
"eval_samples_per_second": 387.418,
|
| 1216 |
+
"eval_steps_per_second": 1.516,
|
| 1217 |
+
"step": 37500
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 0.9310074480595845,
|
| 1221 |
+
"grad_norm": 21.597898483276367,
|
| 1222 |
+
"learning_rate": 2.791694230062477e-05,
|
| 1223 |
+
"loss": 0.7769,
|
| 1224 |
+
"step": 38000
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 0.9310074480595845,
|
| 1228 |
+
"eval_cosine_accuracy": 0.8111377954483032,
|
| 1229 |
+
"eval_loss": 15.210185050964355,
|
| 1230 |
+
"eval_runtime": 49.5904,
|
| 1231 |
+
"eval_samples_per_second": 381.283,
|
| 1232 |
+
"eval_steps_per_second": 1.492,
|
| 1233 |
+
"step": 38000
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 0.9432575460603685,
|
| 1237 |
+
"grad_norm": 9.49044418334961,
|
| 1238 |
+
"learning_rate": 2.8284454244762954e-05,
|
| 1239 |
+
"loss": 0.7848,
|
| 1240 |
+
"step": 38500
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 0.9432575460603685,
|
| 1244 |
+
"eval_cosine_accuracy": 0.8057456016540527,
|
| 1245 |
+
"eval_loss": 14.83116340637207,
|
| 1246 |
+
"eval_runtime": 48.9862,
|
| 1247 |
+
"eval_samples_per_second": 385.986,
|
| 1248 |
+
"eval_steps_per_second": 1.511,
|
| 1249 |
+
"step": 38500
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.9555076440611525,
|
| 1253 |
+
"grad_norm": 4.039228916168213,
|
| 1254 |
+
"learning_rate": 2.8651966188901137e-05,
|
| 1255 |
+
"loss": 0.783,
|
| 1256 |
+
"step": 39000
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.9555076440611525,
|
| 1260 |
+
"eval_cosine_accuracy": 0.8018487691879272,
|
| 1261 |
+
"eval_loss": 15.22596549987793,
|
| 1262 |
+
"eval_runtime": 49.7681,
|
| 1263 |
+
"eval_samples_per_second": 379.922,
|
| 1264 |
+
"eval_steps_per_second": 1.487,
|
| 1265 |
+
"step": 39000
|
| 1266 |
+
},
|
| 1267 |
+
{
|
| 1268 |
+
"epoch": 0.9677577420619365,
|
| 1269 |
+
"grad_norm": 11.418230056762695,
|
| 1270 |
+
"learning_rate": 2.9019478133039323e-05,
|
| 1271 |
+
"loss": 0.7434,
|
| 1272 |
+
"step": 39500
|
| 1273 |
+
},
|
| 1274 |
+
{
|
| 1275 |
+
"epoch": 0.9677577420619365,
|
| 1276 |
+
"eval_cosine_accuracy": 0.8076034188270569,
|
| 1277 |
+
"eval_loss": 15.166121482849121,
|
| 1278 |
+
"eval_runtime": 49.0849,
|
| 1279 |
+
"eval_samples_per_second": 385.21,
|
| 1280 |
+
"eval_steps_per_second": 1.508,
|
| 1281 |
+
"step": 39500
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 0.9800078400627205,
|
| 1285 |
+
"grad_norm": 9.14000415802002,
|
| 1286 |
+
"learning_rate": 2.938699007717751e-05,
|
| 1287 |
+
"loss": 0.7602,
|
| 1288 |
+
"step": 40000
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"epoch": 0.9800078400627205,
|
| 1292 |
+
"eval_cosine_accuracy": 0.7957315444946289,
|
| 1293 |
+
"eval_loss": 15.132609367370605,
|
| 1294 |
+
"eval_runtime": 49.7763,
|
| 1295 |
+
"eval_samples_per_second": 379.86,
|
| 1296 |
+
"eval_steps_per_second": 1.487,
|
| 1297 |
+
"step": 40000
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.9922579380635045,
|
| 1301 |
+
"grad_norm": 10.932486534118652,
|
| 1302 |
+
"learning_rate": 2.9753766997427418e-05,
|
| 1303 |
+
"loss": 0.75,
|
| 1304 |
+
"step": 40500
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.9922579380635045,
|
| 1308 |
+
"eval_cosine_accuracy": 0.7949612736701965,
|
| 1309 |
+
"eval_loss": 15.568105697631836,
|
| 1310 |
+
"eval_runtime": 49.3663,
|
| 1311 |
+
"eval_samples_per_second": 383.014,
|
| 1312 |
+
"eval_steps_per_second": 1.499,
|
| 1313 |
+
"step": 40500
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 1.0045080360642886,
|
| 1317 |
+
"grad_norm": 7.589664459228516,
|
| 1318 |
+
"learning_rate": 2.999986559427645e-05,
|
| 1319 |
+
"loss": 0.7615,
|
| 1320 |
+
"step": 41000
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 1.0045080360642886,
|
| 1324 |
+
"eval_cosine_accuracy": 0.7948253154754639,
|
| 1325 |
+
"eval_loss": 15.345169067382812,
|
| 1326 |
+
"eval_runtime": 49.6024,
|
| 1327 |
+
"eval_samples_per_second": 381.191,
|
| 1328 |
+
"eval_steps_per_second": 1.492,
|
| 1329 |
+
"step": 41000
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 1.0167581340650724,
|
| 1333 |
+
"grad_norm": 6.932123184204102,
|
| 1334 |
+
"learning_rate": 2.9997816851495025e-05,
|
| 1335 |
+
"loss": 0.6745,
|
| 1336 |
+
"step": 41500
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 1.0167581340650724,
|
| 1340 |
+
"eval_cosine_accuracy": 0.7893425226211548,
|
| 1341 |
+
"eval_loss": 15.186988830566406,
|
| 1342 |
+
"eval_runtime": 50.6996,
|
| 1343 |
+
"eval_samples_per_second": 372.942,
|
| 1344 |
+
"eval_steps_per_second": 1.46,
|
| 1345 |
+
"step": 41500
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 1.0290082320658565,
|
| 1349 |
+
"grad_norm": 16.639392852783203,
|
| 1350 |
+
"learning_rate": 2.999330007346105e-05,
|
| 1351 |
+
"loss": 0.6691,
|
| 1352 |
+
"step": 42000
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 1.0290082320658565,
|
| 1356 |
+
"eval_cosine_accuracy": 0.8046128153800964,
|
| 1357 |
+
"eval_loss": 14.84115982055664,
|
| 1358 |
+
"eval_runtime": 49.8193,
|
| 1359 |
+
"eval_samples_per_second": 379.532,
|
| 1360 |
+
"eval_steps_per_second": 1.485,
|
| 1361 |
+
"step": 42000
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.0412583300666405,
|
| 1365 |
+
"grad_norm": 7.760829925537109,
|
| 1366 |
+
"learning_rate": 2.9986316003453863e-05,
|
| 1367 |
+
"loss": 0.7079,
|
| 1368 |
+
"step": 42500
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.0412583300666405,
|
| 1372 |
+
"eval_cosine_accuracy": 0.8026190400123596,
|
| 1373 |
+
"eval_loss": 15.067384719848633,
|
| 1374 |
+
"eval_runtime": 49.3008,
|
| 1375 |
+
"eval_samples_per_second": 383.523,
|
| 1376 |
+
"eval_steps_per_second": 1.501,
|
| 1377 |
+
"step": 42500
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 1.0535084280674245,
|
| 1381 |
+
"grad_norm": 10.852176666259766,
|
| 1382 |
+
"learning_rate": 2.9976865790769467e-05,
|
| 1383 |
+
"loss": 0.6594,
|
| 1384 |
+
"step": 43000
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 1.0535084280674245,
|
| 1388 |
+
"eval_cosine_accuracy": 0.8100050091743469,
|
| 1389 |
+
"eval_loss": 14.875770568847656,
|
| 1390 |
+
"eval_runtime": 49.9156,
|
| 1391 |
+
"eval_samples_per_second": 378.799,
|
| 1392 |
+
"eval_steps_per_second": 1.483,
|
| 1393 |
+
"step": 43000
|
| 1394 |
+
},
|
| 1395 |
+
{
|
| 1396 |
+
"epoch": 1.0657585260682085,
|
| 1397 |
+
"grad_norm": 0.0,
|
| 1398 |
+
"learning_rate": 2.996497727855455e-05,
|
| 1399 |
+
"loss": 0.665,
|
| 1400 |
+
"step": 43500
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 1.0657585260682085,
|
| 1404 |
+
"eval_cosine_accuracy": 0.7897050380706787,
|
| 1405 |
+
"eval_loss": 15.206002235412598,
|
| 1406 |
+
"eval_runtime": 49.028,
|
| 1407 |
+
"eval_samples_per_second": 385.658,
|
| 1408 |
+
"eval_steps_per_second": 1.509,
|
| 1409 |
+
"step": 43500
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 1.0780086240689926,
|
| 1413 |
+
"grad_norm": 10.64773178100586,
|
| 1414 |
+
"learning_rate": 2.9950604774485516e-05,
|
| 1415 |
+
"loss": 0.6751,
|
| 1416 |
+
"step": 44000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 1.0780086240689926,
|
| 1420 |
+
"eval_cosine_accuracy": 0.7938737869262695,
|
| 1421 |
+
"eval_loss": 14.952627182006836,
|
| 1422 |
+
"eval_runtime": 48.8965,
|
| 1423 |
+
"eval_samples_per_second": 386.695,
|
| 1424 |
+
"eval_steps_per_second": 1.513,
|
| 1425 |
+
"step": 44000
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 1.0902587220697766,
|
| 1429 |
+
"grad_norm": 0.0,
|
| 1430 |
+
"learning_rate": 2.993377200436605e-05,
|
| 1431 |
+
"loss": 0.6794,
|
| 1432 |
+
"step": 44500
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 1.0902587220697766,
|
| 1436 |
+
"eval_cosine_accuracy": 0.7745253443717957,
|
| 1437 |
+
"eval_loss": 16.021696090698242,
|
| 1438 |
+
"eval_runtime": 49.7815,
|
| 1439 |
+
"eval_samples_per_second": 379.82,
|
| 1440 |
+
"eval_steps_per_second": 1.486,
|
| 1441 |
+
"step": 44500
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 1.1025088200705606,
|
| 1445 |
+
"grad_norm": 9.036418914794922,
|
| 1446 |
+
"learning_rate": 2.9914481738190634e-05,
|
| 1447 |
+
"loss": 0.6634,
|
| 1448 |
+
"step": 45000
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 1.1025088200705606,
|
| 1452 |
+
"eval_cosine_accuracy": 0.7867597341537476,
|
| 1453 |
+
"eval_loss": 15.794776916503906,
|
| 1454 |
+
"eval_runtime": 48.8265,
|
| 1455 |
+
"eval_samples_per_second": 387.248,
|
| 1456 |
+
"eval_steps_per_second": 1.516,
|
| 1457 |
+
"step": 45000
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 1.1147589180713446,
|
| 1461 |
+
"grad_norm": 5.897200107574463,
|
| 1462 |
+
"learning_rate": 2.989273715035842e-05,
|
| 1463 |
+
"loss": 0.6403,
|
| 1464 |
+
"step": 45500
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 1.1147589180713446,
|
| 1468 |
+
"eval_cosine_accuracy": 0.7909737825393677,
|
| 1469 |
+
"eval_loss": 15.574027061462402,
|
| 1470 |
+
"eval_runtime": 50.269,
|
| 1471 |
+
"eval_samples_per_second": 376.136,
|
| 1472 |
+
"eval_steps_per_second": 1.472,
|
| 1473 |
+
"step": 45500
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.1270090160721287,
|
| 1477 |
+
"grad_norm": 3.216644287109375,
|
| 1478 |
+
"learning_rate": 2.9868541819150817e-05,
|
| 1479 |
+
"loss": 0.662,
|
| 1480 |
+
"step": 46000
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.1270090160721287,
|
| 1484 |
+
"eval_cosine_accuracy": 0.7768362760543823,
|
| 1485 |
+
"eval_loss": 15.986827850341797,
|
| 1486 |
+
"eval_runtime": 49.844,
|
| 1487 |
+
"eval_samples_per_second": 379.343,
|
| 1488 |
+
"eval_steps_per_second": 1.485,
|
| 1489 |
+
"step": 46000
|
| 1490 |
+
},
|
| 1491 |
+
{
|
| 1492 |
+
"epoch": 1.1392591140729125,
|
| 1493 |
+
"grad_norm": 9.330469131469727,
|
| 1494 |
+
"learning_rate": 2.9841899726142677e-05,
|
| 1495 |
+
"loss": 0.6228,
|
| 1496 |
+
"step": 46500
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 1.1392591140729125,
|
| 1500 |
+
"eval_cosine_accuracy": 0.7769269347190857,
|
| 1501 |
+
"eval_loss": 16.077783584594727,
|
| 1502 |
+
"eval_runtime": 49.0491,
|
| 1503 |
+
"eval_samples_per_second": 385.491,
|
| 1504 |
+
"eval_steps_per_second": 1.509,
|
| 1505 |
+
"step": 46500
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 1.1515092120736965,
|
| 1509 |
+
"grad_norm": 13.933734893798828,
|
| 1510 |
+
"learning_rate": 2.9812815255547073e-05,
|
| 1511 |
+
"loss": 0.6524,
|
| 1512 |
+
"step": 47000
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 1.1515092120736965,
|
| 1516 |
+
"eval_cosine_accuracy": 0.7923331260681152,
|
| 1517 |
+
"eval_loss": 15.513272285461426,
|
| 1518 |
+
"eval_runtime": 49.6692,
|
| 1519 |
+
"eval_samples_per_second": 380.679,
|
| 1520 |
+
"eval_steps_per_second": 1.49,
|
| 1521 |
+
"step": 47000
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 1.1637593100744805,
|
| 1525 |
+
"grad_norm": 5.142937183380127,
|
| 1526 |
+
"learning_rate": 2.978135866695313e-05,
|
| 1527 |
+
"loss": 0.5814,
|
| 1528 |
+
"step": 47500
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 1.1637593100744805,
|
| 1532 |
+
"eval_cosine_accuracy": 0.8087815642356873,
|
| 1533 |
+
"eval_loss": 15.18636703491211,
|
| 1534 |
+
"eval_runtime": 49.1431,
|
| 1535 |
+
"eval_samples_per_second": 384.754,
|
| 1536 |
+
"eval_steps_per_second": 1.506,
|
| 1537 |
+
"step": 47500
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 1.1760094080752646,
|
| 1541 |
+
"grad_norm": 9.375741958618164,
|
| 1542 |
+
"learning_rate": 2.9747409060066586e-05,
|
| 1543 |
+
"loss": 0.6097,
|
| 1544 |
+
"step": 48000
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 1.1760094080752646,
|
| 1548 |
+
"eval_cosine_accuracy": 0.7749331593513489,
|
| 1549 |
+
"eval_loss": 15.985566139221191,
|
| 1550 |
+
"eval_runtime": 48.8698,
|
| 1551 |
+
"eval_samples_per_second": 386.906,
|
| 1552 |
+
"eval_steps_per_second": 1.514,
|
| 1553 |
+
"step": 48000
|
| 1554 |
+
},
|
| 1555 |
+
{
|
| 1556 |
+
"epoch": 1.1882595060760486,
|
| 1557 |
+
"grad_norm": 8.774680137634277,
|
| 1558 |
+
"learning_rate": 2.971103262494201e-05,
|
| 1559 |
+
"loss": 0.6098,
|
| 1560 |
+
"step": 48500
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 1.1882595060760486,
|
| 1564 |
+
"eval_cosine_accuracy": 0.7782862782478333,
|
| 1565 |
+
"eval_loss": 15.625438690185547,
|
| 1566 |
+
"eval_runtime": 50.0734,
|
| 1567 |
+
"eval_samples_per_second": 377.606,
|
| 1568 |
+
"eval_steps_per_second": 1.478,
|
| 1569 |
+
"step": 48500
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 1.2005096040768326,
|
| 1573 |
+
"grad_norm": 13.765412330627441,
|
| 1574 |
+
"learning_rate": 2.967223534767226e-05,
|
| 1575 |
+
"loss": 0.6019,
|
| 1576 |
+
"step": 49000
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 1.2005096040768326,
|
| 1580 |
+
"eval_cosine_accuracy": 0.7873034477233887,
|
| 1581 |
+
"eval_loss": 15.482627868652344,
|
| 1582 |
+
"eval_runtime": 49.8059,
|
| 1583 |
+
"eval_samples_per_second": 379.633,
|
| 1584 |
+
"eval_steps_per_second": 1.486,
|
| 1585 |
+
"step": 49000
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.2127597020776166,
|
| 1589 |
+
"grad_norm": 4.794047832489014,
|
| 1590 |
+
"learning_rate": 2.9631108441380145e-05,
|
| 1591 |
+
"loss": 0.5903,
|
| 1592 |
+
"step": 49500
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.2127597020776166,
|
| 1596 |
+
"eval_cosine_accuracy": 0.7844034433364868,
|
| 1597 |
+
"eval_loss": 15.559972763061523,
|
| 1598 |
+
"eval_runtime": 48.9919,
|
| 1599 |
+
"eval_samples_per_second": 385.941,
|
| 1600 |
+
"eval_steps_per_second": 1.51,
|
| 1601 |
+
"step": 49500
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"epoch": 1.2250098000784007,
|
| 1605 |
+
"grad_norm": 8.011475563049316,
|
| 1606 |
+
"learning_rate": 2.958749383885957e-05,
|
| 1607 |
+
"loss": 0.5999,
|
| 1608 |
+
"step": 50000
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 1.2250098000784007,
|
| 1612 |
+
"eval_cosine_accuracy": 0.7828628420829773,
|
| 1613 |
+
"eval_loss": 15.543310165405273,
|
| 1614 |
+
"eval_runtime": 49.9668,
|
| 1615 |
+
"eval_samples_per_second": 378.411,
|
| 1616 |
+
"eval_steps_per_second": 1.481,
|
| 1617 |
+
"step": 50000
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 1.2372598980791847,
|
| 1621 |
+
"grad_norm": 12.538202285766602,
|
| 1622 |
+
"learning_rate": 2.9541478723691243e-05,
|
| 1623 |
+
"loss": 0.5942,
|
| 1624 |
+
"step": 50500
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"epoch": 1.2372598980791847,
|
| 1628 |
+
"eval_cosine_accuracy": 0.7695409655570984,
|
| 1629 |
+
"eval_loss": 15.88884162902832,
|
| 1630 |
+
"eval_runtime": 49.0955,
|
| 1631 |
+
"eval_samples_per_second": 385.127,
|
| 1632 |
+
"eval_steps_per_second": 1.507,
|
| 1633 |
+
"step": 50500
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 1.2495099960799687,
|
| 1637 |
+
"grad_norm": 8.51110553741455,
|
| 1638 |
+
"learning_rate": 2.949307066810568e-05,
|
| 1639 |
+
"loss": 0.5658,
|
| 1640 |
+
"step": 51000
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 1.2495099960799687,
|
| 1644 |
+
"eval_cosine_accuracy": 0.7802800536155701,
|
| 1645 |
+
"eval_loss": 15.716775894165039,
|
| 1646 |
+
"eval_runtime": 49.729,
|
| 1647 |
+
"eval_samples_per_second": 380.221,
|
| 1648 |
+
"eval_steps_per_second": 1.488,
|
| 1649 |
+
"step": 51000
|
| 1650 |
+
},
|
| 1651 |
+
{
|
| 1652 |
+
"epoch": 1.2617600940807527,
|
| 1653 |
+
"grad_norm": 13.569243431091309,
|
| 1654 |
+
"learning_rate": 2.9442277638114786e-05,
|
| 1655 |
+
"loss": 0.5518,
|
| 1656 |
+
"step": 51500
|
| 1657 |
+
},
|
| 1658 |
+
{
|
| 1659 |
+
"epoch": 1.2617600940807527,
|
| 1660 |
+
"eval_cosine_accuracy": 0.7801893949508667,
|
| 1661 |
+
"eval_loss": 15.803173065185547,
|
| 1662 |
+
"eval_runtime": 48.9528,
|
| 1663 |
+
"eval_samples_per_second": 386.249,
|
| 1664 |
+
"eval_steps_per_second": 1.512,
|
| 1665 |
+
"step": 51500
|
| 1666 |
+
},
|
| 1667 |
+
{
|
| 1668 |
+
"epoch": 1.2740101920815365,
|
| 1669 |
+
"grad_norm": 7.125584602355957,
|
| 1670 |
+
"learning_rate": 2.9389107992200996e-05,
|
| 1671 |
+
"loss": 0.5632,
|
| 1672 |
+
"step": 52000
|
| 1673 |
+
},
|
| 1674 |
+
{
|
| 1675 |
+
"epoch": 1.2740101920815365,
|
| 1676 |
+
"eval_cosine_accuracy": 0.7831346988677979,
|
| 1677 |
+
"eval_loss": 16.11168670654297,
|
| 1678 |
+
"eval_runtime": 49.8742,
|
| 1679 |
+
"eval_samples_per_second": 379.114,
|
| 1680 |
+
"eval_steps_per_second": 1.484,
|
| 1681 |
+
"step": 52000
|
| 1682 |
+
},
|
| 1683 |
+
{
|
| 1684 |
+
"epoch": 1.2862602900823208,
|
| 1685 |
+
"grad_norm": 0.0,
|
| 1686 |
+
"learning_rate": 2.9333570479941778e-05,
|
| 1687 |
+
"loss": 0.5576,
|
| 1688 |
+
"step": 52500
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 1.2862602900823208,
|
| 1692 |
+
"eval_cosine_accuracy": 0.7808237671852112,
|
| 1693 |
+
"eval_loss": 16.117074966430664,
|
| 1694 |
+
"eval_runtime": 49.071,
|
| 1695 |
+
"eval_samples_per_second": 385.319,
|
| 1696 |
+
"eval_steps_per_second": 1.508,
|
| 1697 |
+
"step": 52500
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 1.2985103880831046,
|
| 1701 |
+
"grad_norm": 10.205459594726562,
|
| 1702 |
+
"learning_rate": 2.9275674240569822e-05,
|
| 1703 |
+
"loss": 0.5558,
|
| 1704 |
+
"step": 53000
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 1.2985103880831046,
|
| 1708 |
+
"eval_cosine_accuracy": 0.7743440866470337,
|
| 1709 |
+
"eval_loss": 15.681249618530273,
|
| 1710 |
+
"eval_runtime": 49.9495,
|
| 1711 |
+
"eval_samples_per_second": 378.543,
|
| 1712 |
+
"eval_steps_per_second": 1.481,
|
| 1713 |
+
"step": 53000
|
| 1714 |
+
},
|
| 1715 |
+
{
|
| 1716 |
+
"epoch": 1.3107604860838886,
|
| 1717 |
+
"grad_norm": 7.976107597351074,
|
| 1718 |
+
"learning_rate": 2.9215551630323343e-05,
|
| 1719 |
+
"loss": 0.5644,
|
| 1720 |
+
"step": 53500
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 1.3107604860838886,
|
| 1724 |
+
"eval_cosine_accuracy": 0.764194130897522,
|
| 1725 |
+
"eval_loss": 15.613662719726562,
|
| 1726 |
+
"eval_runtime": 49.1596,
|
| 1727 |
+
"eval_samples_per_second": 384.625,
|
| 1728 |
+
"eval_steps_per_second": 1.505,
|
| 1729 |
+
"step": 53500
|
| 1730 |
+
},
|
| 1731 |
+
{
|
| 1732 |
+
"epoch": 1.3230105840846726,
|
| 1733 |
+
"grad_norm": 11.447208404541016,
|
| 1734 |
+
"learning_rate": 2.9153099061841822e-05,
|
| 1735 |
+
"loss": 0.5505,
|
| 1736 |
+
"step": 54000
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 1.3230105840846726,
|
| 1740 |
+
"eval_cosine_accuracy": 0.7777425050735474,
|
| 1741 |
+
"eval_loss": 15.998211860656738,
|
| 1742 |
+
"eval_runtime": 49.0768,
|
| 1743 |
+
"eval_samples_per_second": 385.273,
|
| 1744 |
+
"eval_steps_per_second": 1.508,
|
| 1745 |
+
"step": 54000
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 1.3352606820854567,
|
| 1749 |
+
"grad_norm": 0.0,
|
| 1750 |
+
"learning_rate": 2.9088194645061204e-05,
|
| 1751 |
+
"loss": 0.5812,
|
| 1752 |
+
"step": 54500
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 1.3352606820854567,
|
| 1756 |
+
"eval_cosine_accuracy": 0.7656894326210022,
|
| 1757 |
+
"eval_loss": 16.415210723876953,
|
| 1758 |
+
"eval_runtime": 50.4064,
|
| 1759 |
+
"eval_samples_per_second": 375.111,
|
| 1760 |
+
"eval_steps_per_second": 1.468,
|
| 1761 |
+
"step": 54500
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 1.3475107800862407,
|
| 1765 |
+
"grad_norm": 3.6030280590057373,
|
| 1766 |
+
"learning_rate": 2.902097188012544e-05,
|
| 1767 |
+
"loss": 0.5423,
|
| 1768 |
+
"step": 55000
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"epoch": 1.3475107800862407,
|
| 1772 |
+
"eval_cosine_accuracy": 0.766187846660614,
|
| 1773 |
+
"eval_loss": 15.649203300476074,
|
| 1774 |
+
"eval_runtime": 49.0755,
|
| 1775 |
+
"eval_samples_per_second": 385.284,
|
| 1776 |
+
"eval_steps_per_second": 1.508,
|
| 1777 |
+
"step": 55000
|
| 1778 |
+
},
|
| 1779 |
+
{
|
| 1780 |
+
"epoch": 1.3597608780870247,
|
| 1781 |
+
"grad_norm": 6.679117202758789,
|
| 1782 |
+
"learning_rate": 2.8951441829188132e-05,
|
| 1783 |
+
"loss": 0.557,
|
| 1784 |
+
"step": 55500
|
| 1785 |
+
},
|
| 1786 |
+
{
|
| 1787 |
+
"epoch": 1.3597608780870247,
|
| 1788 |
+
"eval_cosine_accuracy": 0.7639675736427307,
|
| 1789 |
+
"eval_loss": 16.437679290771484,
|
| 1790 |
+
"eval_runtime": 50.2106,
|
| 1791 |
+
"eval_samples_per_second": 376.574,
|
| 1792 |
+
"eval_steps_per_second": 1.474,
|
| 1793 |
+
"step": 55500
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 1.3720109760878088,
|
| 1797 |
+
"grad_norm": 0.0,
|
| 1798 |
+
"learning_rate": 2.887961593408902e-05,
|
| 1799 |
+
"loss": 0.5419,
|
| 1800 |
+
"step": 56000
|
| 1801 |
+
},
|
| 1802 |
+
{
|
| 1803 |
+
"epoch": 1.3720109760878088,
|
| 1804 |
+
"eval_cosine_accuracy": 0.7617019414901733,
|
| 1805 |
+
"eval_loss": 16.5505313873291,
|
| 1806 |
+
"eval_runtime": 49.4643,
|
| 1807 |
+
"eval_samples_per_second": 382.256,
|
| 1808 |
+
"eval_steps_per_second": 1.496,
|
| 1809 |
+
"step": 56000
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 1.3842610740885928,
|
| 1813 |
+
"grad_norm": 7.8234734535217285,
|
| 1814 |
+
"learning_rate": 2.880550601447111e-05,
|
| 1815 |
+
"loss": 0.5178,
|
| 1816 |
+
"step": 56500
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 1.3842610740885928,
|
| 1820 |
+
"eval_cosine_accuracy": 0.7699940800666809,
|
| 1821 |
+
"eval_loss": 16.079374313354492,
|
| 1822 |
+
"eval_runtime": 48.8743,
|
| 1823 |
+
"eval_samples_per_second": 386.87,
|
| 1824 |
+
"eval_steps_per_second": 1.514,
|
| 1825 |
+
"step": 56500
|
| 1826 |
+
},
|
| 1827 |
+
{
|
| 1828 |
+
"epoch": 1.3965111720893768,
|
| 1829 |
+
"grad_norm": 9.114197731018066,
|
| 1830 |
+
"learning_rate": 2.8729124265835657e-05,
|
| 1831 |
+
"loss": 0.5385,
|
| 1832 |
+
"step": 57000
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"epoch": 1.3965111720893768,
|
| 1836 |
+
"eval_cosine_accuracy": 0.7776972055435181,
|
| 1837 |
+
"eval_loss": 15.54992961883545,
|
| 1838 |
+
"eval_runtime": 49.558,
|
| 1839 |
+
"eval_samples_per_second": 381.533,
|
| 1840 |
+
"eval_steps_per_second": 1.493,
|
| 1841 |
+
"step": 57000
|
| 1842 |
+
},
|
| 1843 |
+
{
|
| 1844 |
+
"epoch": 1.4087612700901606,
|
| 1845 |
+
"grad_norm": 6.118236541748047,
|
| 1846 |
+
"learning_rate": 2.8650483257535268e-05,
|
| 1847 |
+
"loss": 0.5384,
|
| 1848 |
+
"step": 57500
|
| 1849 |
+
},
|
| 1850 |
+
{
|
| 1851 |
+
"epoch": 1.4087612700901606,
|
| 1852 |
+
"eval_cosine_accuracy": 0.7788300514221191,
|
| 1853 |
+
"eval_loss": 15.954768180847168,
|
| 1854 |
+
"eval_runtime": 48.867,
|
| 1855 |
+
"eval_samples_per_second": 386.928,
|
| 1856 |
+
"eval_steps_per_second": 1.514,
|
| 1857 |
+
"step": 57500
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 1.4210113680909449,
|
| 1861 |
+
"grad_norm": 12.296391487121582,
|
| 1862 |
+
"learning_rate": 2.8569759938399108e-05,
|
| 1863 |
+
"loss": 0.52,
|
| 1864 |
+
"step": 58000
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 1.4210113680909449,
|
| 1868 |
+
"eval_cosine_accuracy": 0.7652363181114197,
|
| 1869 |
+
"eval_loss": 16.53635597229004,
|
| 1870 |
+
"eval_runtime": 49.6444,
|
| 1871 |
+
"eval_samples_per_second": 380.868,
|
| 1872 |
+
"eval_steps_per_second": 1.491,
|
| 1873 |
+
"step": 58000
|
| 1874 |
+
},
|
| 1875 |
+
{
|
| 1876 |
+
"epoch": 1.4332614660917287,
|
| 1877 |
+
"grad_norm": 5.053269386291504,
|
| 1878 |
+
"learning_rate": 2.8486644056315534e-05,
|
| 1879 |
+
"loss": 0.5155,
|
| 1880 |
+
"step": 58500
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 1.4332614660917287,
|
| 1884 |
+
"eval_cosine_accuracy": 0.7767003774642944,
|
| 1885 |
+
"eval_loss": 15.921165466308594,
|
| 1886 |
+
"eval_runtime": 49.1556,
|
| 1887 |
+
"eval_samples_per_second": 384.656,
|
| 1888 |
+
"eval_steps_per_second": 1.505,
|
| 1889 |
+
"step": 58500
|
| 1890 |
+
},
|
| 1891 |
+
{
|
| 1892 |
+
"epoch": 1.4455115640925127,
|
| 1893 |
+
"grad_norm": 0.0,
|
| 1894 |
+
"learning_rate": 2.8401308817021564e-05,
|
| 1895 |
+
"loss": 0.5132,
|
| 1896 |
+
"step": 59000
|
| 1897 |
+
},
|
| 1898 |
+
{
|
| 1899 |
+
"epoch": 1.4455115640925127,
|
| 1900 |
+
"eval_cosine_accuracy": 0.7660519480705261,
|
| 1901 |
+
"eval_loss": 16.090425491333008,
|
| 1902 |
+
"eval_runtime": 49.0018,
|
| 1903 |
+
"eval_samples_per_second": 385.864,
|
| 1904 |
+
"eval_steps_per_second": 1.51,
|
| 1905 |
+
"step": 59000
|
| 1906 |
+
},
|
| 1907 |
+
{
|
| 1908 |
+
"epoch": 1.4577616620932967,
|
| 1909 |
+
"grad_norm": 5.571829319000244,
|
| 1910 |
+
"learning_rate": 2.8313768263252943e-05,
|
| 1911 |
+
"loss": 0.541,
|
| 1912 |
+
"step": 59500
|
| 1913 |
+
},
|
| 1914 |
+
{
|
| 1915 |
+
"epoch": 1.4577616620932967,
|
| 1916 |
+
"eval_cosine_accuracy": 0.7597535252571106,
|
| 1917 |
+
"eval_loss": 16.565155029296875,
|
| 1918 |
+
"eval_runtime": 50.3529,
|
| 1919 |
+
"eval_samples_per_second": 375.51,
|
| 1920 |
+
"eval_steps_per_second": 1.47,
|
| 1921 |
+
"step": 59500
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 1.4700117600940807,
|
| 1925 |
+
"grad_norm": 10.40772533416748,
|
| 1926 |
+
"learning_rate": 2.8224218440348566e-05,
|
| 1927 |
+
"loss": 0.5159,
|
| 1928 |
+
"step": 60000
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 1.4700117600940807,
|
| 1932 |
+
"eval_cosine_accuracy": 0.7472019791603088,
|
| 1933 |
+
"eval_loss": 16.681304931640625,
|
| 1934 |
+
"eval_runtime": 49.7085,
|
| 1935 |
+
"eval_samples_per_second": 380.378,
|
| 1936 |
+
"eval_steps_per_second": 1.489,
|
| 1937 |
+
"step": 60000
|
| 1938 |
+
},
|
| 1939 |
+
{
|
| 1940 |
+
"epoch": 1.4822618580948648,
|
| 1941 |
+
"grad_norm": 7.996941566467285,
|
| 1942 |
+
"learning_rate": 2.8132315172400515e-05,
|
| 1943 |
+
"loss": 0.5315,
|
| 1944 |
+
"step": 60500
|
| 1945 |
+
},
|
| 1946 |
+
{
|
| 1947 |
+
"epoch": 1.4822618580948648,
|
| 1948 |
+
"eval_cosine_accuracy": 0.7678191065788269,
|
| 1949 |
+
"eval_loss": 16.14435577392578,
|
| 1950 |
+
"eval_runtime": 48.725,
|
| 1951 |
+
"eval_samples_per_second": 388.056,
|
| 1952 |
+
"eval_steps_per_second": 1.519,
|
| 1953 |
+
"step": 60500
|
| 1954 |
+
},
|
| 1955 |
+
{
|
| 1956 |
+
"epoch": 1.4945119560956488,
|
| 1957 |
+
"grad_norm": 0.0,
|
| 1958 |
+
"learning_rate": 2.8038250855473538e-05,
|
| 1959 |
+
"loss": 0.5084,
|
| 1960 |
+
"step": 61000
|
| 1961 |
+
},
|
| 1962 |
+
{
|
| 1963 |
+
"epoch": 1.4945119560956488,
|
| 1964 |
+
"eval_cosine_accuracy": 0.7491050958633423,
|
| 1965 |
+
"eval_loss": 16.488277435302734,
|
| 1966 |
+
"eval_runtime": 49.8772,
|
| 1967 |
+
"eval_samples_per_second": 379.091,
|
| 1968 |
+
"eval_steps_per_second": 1.484,
|
| 1969 |
+
"step": 61000
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 1.5067620540964328,
|
| 1973 |
+
"grad_norm": 10.695802688598633,
|
| 1974 |
+
"learning_rate": 2.7942040968757193e-05,
|
| 1975 |
+
"loss": 0.484,
|
| 1976 |
+
"step": 61500
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 1.5067620540964328,
|
| 1980 |
+
"eval_cosine_accuracy": 0.7501019239425659,
|
| 1981 |
+
"eval_loss": 16.454145431518555,
|
| 1982 |
+
"eval_runtime": 48.7343,
|
| 1983 |
+
"eval_samples_per_second": 387.981,
|
| 1984 |
+
"eval_steps_per_second": 1.518,
|
| 1985 |
+
"step": 61500
|
| 1986 |
+
},
|
| 1987 |
+
{
|
| 1988 |
+
"epoch": 1.5190121520972166,
|
| 1989 |
+
"grad_norm": 7.5936598777771,
|
| 1990 |
+
"learning_rate": 2.7843701344515244e-05,
|
| 1991 |
+
"loss": 0.4808,
|
| 1992 |
+
"step": 62000
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 1.5190121520972166,
|
| 1996 |
+
"eval_cosine_accuracy": 0.751053512096405,
|
| 1997 |
+
"eval_loss": 16.385787963867188,
|
| 1998 |
+
"eval_runtime": 50.1864,
|
| 1999 |
+
"eval_samples_per_second": 376.755,
|
| 2000 |
+
"eval_steps_per_second": 1.475,
|
| 2001 |
+
"step": 62000
|
| 2002 |
+
},
|
| 2003 |
+
{
|
| 2004 |
+
"epoch": 1.5312622500980009,
|
| 2005 |
+
"grad_norm": 9.096808433532715,
|
| 2006 |
+
"learning_rate": 2.7743654166604067e-05,
|
| 2007 |
+
"loss": 0.4893,
|
| 2008 |
+
"step": 62500
|
| 2009 |
+
},
|
| 2010 |
+
{
|
| 2011 |
+
"epoch": 1.5312622500980009,
|
| 2012 |
+
"eval_cosine_accuracy": 0.7569894194602966,
|
| 2013 |
+
"eval_loss": 16.61661720275879,
|
| 2014 |
+
"eval_runtime": 48.8868,
|
| 2015 |
+
"eval_samples_per_second": 386.771,
|
| 2016 |
+
"eval_steps_per_second": 1.514,
|
| 2017 |
+
"step": 62500
|
| 2018 |
+
},
|
| 2019 |
+
{
|
| 2020 |
+
"epoch": 1.5435123480987847,
|
| 2021 |
+
"grad_norm": 14.026432991027832,
|
| 2022 |
+
"learning_rate": 2.7641112318024342e-05,
|
| 2023 |
+
"loss": 0.5022,
|
| 2024 |
+
"step": 63000
|
| 2025 |
+
},
|
| 2026 |
+
{
|
| 2027 |
+
"epoch": 1.5435123480987847,
|
| 2028 |
+
"eval_cosine_accuracy": 0.7584394216537476,
|
| 2029 |
+
"eval_loss": 16.342727661132812,
|
| 2030 |
+
"eval_runtime": 48.8025,
|
| 2031 |
+
"eval_samples_per_second": 387.439,
|
| 2032 |
+
"eval_steps_per_second": 1.516,
|
| 2033 |
+
"step": 63000
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 1.555762446099569,
|
| 2037 |
+
"grad_norm": 8.053553581237793,
|
| 2038 |
+
"learning_rate": 2.753649025262765e-05,
|
| 2039 |
+
"loss": 0.4992,
|
| 2040 |
+
"step": 63500
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 1.555762446099569,
|
| 2044 |
+
"eval_cosine_accuracy": 0.7613394260406494,
|
| 2045 |
+
"eval_loss": 16.419708251953125,
|
| 2046 |
+
"eval_runtime": 50.711,
|
| 2047 |
+
"eval_samples_per_second": 372.858,
|
| 2048 |
+
"eval_steps_per_second": 1.459,
|
| 2049 |
+
"step": 63500
|
| 2050 |
+
},
|
| 2051 |
+
{
|
| 2052 |
+
"epoch": 1.5680125441003527,
|
| 2053 |
+
"grad_norm": 5.76508092880249,
|
| 2054 |
+
"learning_rate": 2.742980518698278e-05,
|
| 2055 |
+
"loss": 0.4787,
|
| 2056 |
+
"step": 64000
|
| 2057 |
+
},
|
| 2058 |
+
{
|
| 2059 |
+
"epoch": 1.5680125441003527,
|
| 2060 |
+
"eval_cosine_accuracy": 0.7609316110610962,
|
| 2061 |
+
"eval_loss": 15.770021438598633,
|
| 2062 |
+
"eval_runtime": 49.4659,
|
| 2063 |
+
"eval_samples_per_second": 382.243,
|
| 2064 |
+
"eval_steps_per_second": 1.496,
|
| 2065 |
+
"step": 64000
|
| 2066 |
+
},
|
| 2067 |
+
{
|
| 2068 |
+
"epoch": 1.5802626421011368,
|
| 2069 |
+
"grad_norm": 12.403212547302246,
|
| 2070 |
+
"learning_rate": 2.732107467714509e-05,
|
| 2071 |
+
"loss": 0.4673,
|
| 2072 |
+
"step": 64500
|
| 2073 |
+
},
|
| 2074 |
+
{
|
| 2075 |
+
"epoch": 1.5802626421011368,
|
| 2076 |
+
"eval_cosine_accuracy": 0.752503514289856,
|
| 2077 |
+
"eval_loss": 16.45221519470215,
|
| 2078 |
+
"eval_runtime": 48.9688,
|
| 2079 |
+
"eval_samples_per_second": 386.124,
|
| 2080 |
+
"eval_steps_per_second": 1.511,
|
| 2081 |
+
"step": 64500
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 1.5925127401019208,
|
| 2085 |
+
"grad_norm": 0.0,
|
| 2086 |
+
"learning_rate": 2.721031661576749e-05,
|
| 2087 |
+
"loss": 0.497,
|
| 2088 |
+
"step": 65000
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 1.5925127401019208,
|
| 2092 |
+
"eval_cosine_accuracy": 0.7553128600120544,
|
| 2093 |
+
"eval_loss": 16.404518127441406,
|
| 2094 |
+
"eval_runtime": 49.5893,
|
| 2095 |
+
"eval_samples_per_second": 381.292,
|
| 2096 |
+
"eval_steps_per_second": 1.492,
|
| 2097 |
+
"step": 65000
|
| 2098 |
+
},
|
| 2099 |
+
{
|
| 2100 |
+
"epoch": 1.6047628381027048,
|
| 2101 |
+
"grad_norm": 2.619753122329712,
|
| 2102 |
+
"learning_rate": 2.7097549229156036e-05,
|
| 2103 |
+
"loss": 0.4678,
|
| 2104 |
+
"step": 65500
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"epoch": 1.6047628381027048,
|
| 2108 |
+
"eval_cosine_accuracy": 0.7507363557815552,
|
| 2109 |
+
"eval_loss": 16.57503890991211,
|
| 2110 |
+
"eval_runtime": 48.9192,
|
| 2111 |
+
"eval_samples_per_second": 386.515,
|
| 2112 |
+
"eval_steps_per_second": 1.513,
|
| 2113 |
+
"step": 65500
|
| 2114 |
+
},
|
| 2115 |
+
{
|
| 2116 |
+
"epoch": 1.6170129361034888,
|
| 2117 |
+
"grad_norm": 10.215866088867188,
|
| 2118 |
+
"learning_rate": 2.6982791074270597e-05,
|
| 2119 |
+
"loss": 0.4655,
|
| 2120 |
+
"step": 66000
|
| 2121 |
+
},
|
| 2122 |
+
{
|
| 2123 |
+
"epoch": 1.6170129361034888,
|
| 2124 |
+
"eval_cosine_accuracy": 0.7636956572532654,
|
| 2125 |
+
"eval_loss": 16.23915672302246,
|
| 2126 |
+
"eval_runtime": 49.0341,
|
| 2127 |
+
"eval_samples_per_second": 385.61,
|
| 2128 |
+
"eval_steps_per_second": 1.509,
|
| 2129 |
+
"step": 66000
|
| 2130 |
+
},
|
| 2131 |
+
{
|
| 2132 |
+
"epoch": 1.6292630341042729,
|
| 2133 |
+
"grad_norm": 6.002055644989014,
|
| 2134 |
+
"learning_rate": 2.6866061035671133e-05,
|
| 2135 |
+
"loss": 0.4593,
|
| 2136 |
+
"step": 66500
|
| 2137 |
+
},
|
| 2138 |
+
{
|
| 2139 |
+
"epoch": 1.6292630341042729,
|
| 2140 |
+
"eval_cosine_accuracy": 0.7583941221237183,
|
| 2141 |
+
"eval_loss": 16.42445945739746,
|
| 2142 |
+
"eval_runtime": 50.2733,
|
| 2143 |
+
"eval_samples_per_second": 376.104,
|
| 2144 |
+
"eval_steps_per_second": 1.472,
|
| 2145 |
+
"step": 66500
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 1.641513132105057,
|
| 2149 |
+
"grad_norm": 5.777263164520264,
|
| 2150 |
+
"learning_rate": 2.674737832241009e-05,
|
| 2151 |
+
"loss": 0.445,
|
| 2152 |
+
"step": 67000
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 1.641513132105057,
|
| 2156 |
+
"eval_cosine_accuracy": 0.7737550139427185,
|
| 2157 |
+
"eval_loss": 15.90536880493164,
|
| 2158 |
+
"eval_runtime": 49.6298,
|
| 2159 |
+
"eval_samples_per_second": 380.981,
|
| 2160 |
+
"eval_steps_per_second": 1.491,
|
| 2161 |
+
"step": 67000
|
| 2162 |
+
},
|
| 2163 |
+
{
|
| 2164 |
+
"epoch": 1.6537632301058407,
|
| 2165 |
+
"grad_norm": 11.085003852844238,
|
| 2166 |
+
"learning_rate": 2.662700561272437e-05,
|
| 2167 |
+
"loss": 0.4531,
|
| 2168 |
+
"step": 67500
|
| 2169 |
+
},
|
| 2170 |
+
{
|
| 2171 |
+
"epoch": 1.6537632301058407,
|
| 2172 |
+
"eval_cosine_accuracy": 0.7579410076141357,
|
| 2173 |
+
"eval_loss": 16.559391021728516,
|
| 2174 |
+
"eval_runtime": 49.141,
|
| 2175 |
+
"eval_samples_per_second": 384.771,
|
| 2176 |
+
"eval_steps_per_second": 1.506,
|
| 2177 |
+
"step": 67500
|
| 2178 |
+
},
|
| 2179 |
+
{
|
| 2180 |
+
"epoch": 1.666013328106625,
|
| 2181 |
+
"grad_norm": 10.818819046020508,
|
| 2182 |
+
"learning_rate": 2.6504480265982195e-05,
|
| 2183 |
+
"loss": 0.4413,
|
| 2184 |
+
"step": 68000
|
| 2185 |
+
},
|
| 2186 |
+
{
|
| 2187 |
+
"epoch": 1.666013328106625,
|
| 2188 |
+
"eval_cosine_accuracy": 0.763106644153595,
|
| 2189 |
+
"eval_loss": 16.645488739013672,
|
| 2190 |
+
"eval_runtime": 49.5731,
|
| 2191 |
+
"eval_samples_per_second": 381.417,
|
| 2192 |
+
"eval_steps_per_second": 1.493,
|
| 2193 |
+
"step": 68000
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 1.6782634261074088,
|
| 2197 |
+
"grad_norm": 9.073834419250488,
|
| 2198 |
+
"learning_rate": 2.638006174617757e-05,
|
| 2199 |
+
"loss": 0.4325,
|
| 2200 |
+
"step": 68500
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 1.6782634261074088,
|
| 2204 |
+
"eval_cosine_accuracy": 0.7497847676277161,
|
| 2205 |
+
"eval_loss": 16.888395309448242,
|
| 2206 |
+
"eval_runtime": 49.0193,
|
| 2207 |
+
"eval_samples_per_second": 385.726,
|
| 2208 |
+
"eval_steps_per_second": 1.51,
|
| 2209 |
+
"step": 68500
|
| 2210 |
+
},
|
| 2211 |
+
{
|
| 2212 |
+
"epoch": 1.690513524108193,
|
| 2213 |
+
"grad_norm": 4.124987602233887,
|
| 2214 |
+
"learning_rate": 2.625377052757659e-05,
|
| 2215 |
+
"loss": 0.446,
|
| 2216 |
+
"step": 69000
|
| 2217 |
+
},
|
| 2218 |
+
{
|
| 2219 |
+
"epoch": 1.690513524108193,
|
| 2220 |
+
"eval_cosine_accuracy": 0.7763378620147705,
|
| 2221 |
+
"eval_loss": 16.116737365722656,
|
| 2222 |
+
"eval_runtime": 49.8224,
|
| 2223 |
+
"eval_samples_per_second": 379.508,
|
| 2224 |
+
"eval_steps_per_second": 1.485,
|
| 2225 |
+
"step": 69000
|
| 2226 |
+
},
|
| 2227 |
+
{
|
| 2228 |
+
"epoch": 1.7027636221089768,
|
| 2229 |
+
"grad_norm": 3.8445913791656494,
|
| 2230 |
+
"learning_rate": 2.6125885513132833e-05,
|
| 2231 |
+
"loss": 0.4306,
|
| 2232 |
+
"step": 69500
|
| 2233 |
+
},
|
| 2234 |
+
{
|
| 2235 |
+
"epoch": 1.7027636221089768,
|
| 2236 |
+
"eval_cosine_accuracy": 0.7625628709793091,
|
| 2237 |
+
"eval_loss": 16.68659019470215,
|
| 2238 |
+
"eval_runtime": 49.1194,
|
| 2239 |
+
"eval_samples_per_second": 384.94,
|
| 2240 |
+
"eval_steps_per_second": 1.507,
|
| 2241 |
+
"step": 69500
|
| 2242 |
+
},
|
| 2243 |
+
{
|
| 2244 |
+
"epoch": 1.7150137201097608,
|
| 2245 |
+
"grad_norm": 0.0,
|
| 2246 |
+
"learning_rate": 2.5995915189412465e-05,
|
| 2247 |
+
"loss": 0.4599,
|
| 2248 |
+
"step": 70000
|
| 2249 |
+
},
|
| 2250 |
+
{
|
| 2251 |
+
"epoch": 1.7150137201097608,
|
| 2252 |
+
"eval_cosine_accuracy": 0.7486066222190857,
|
| 2253 |
+
"eval_loss": 16.457317352294922,
|
| 2254 |
+
"eval_runtime": 49.739,
|
| 2255 |
+
"eval_samples_per_second": 380.144,
|
| 2256 |
+
"eval_steps_per_second": 1.488,
|
| 2257 |
+
"step": 70000
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 1.7272638181105449,
|
| 2261 |
+
"grad_norm": 6.720544338226318,
|
| 2262 |
+
"learning_rate": 2.5864135381912537e-05,
|
| 2263 |
+
"loss": 0.423,
|
| 2264 |
+
"step": 70500
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 1.7272638181105449,
|
| 2268 |
+
"eval_cosine_accuracy": 0.7486066222190857,
|
| 2269 |
+
"eval_loss": 16.359926223754883,
|
| 2270 |
+
"eval_runtime": 48.9112,
|
| 2271 |
+
"eval_samples_per_second": 386.578,
|
| 2272 |
+
"eval_steps_per_second": 1.513,
|
| 2273 |
+
"step": 70500
|
| 2274 |
+
},
|
| 2275 |
+
{
|
| 2276 |
+
"epoch": 1.7395139161113289,
|
| 2277 |
+
"grad_norm": 7.174712657928467,
|
| 2278 |
+
"learning_rate": 2.5730567776269952e-05,
|
| 2279 |
+
"loss": 0.4248,
|
| 2280 |
+
"step": 71000
|
| 2281 |
+
},
|
| 2282 |
+
{
|
| 2283 |
+
"epoch": 1.7395139161113289,
|
| 2284 |
+
"eval_cosine_accuracy": 0.7649644017219543,
|
| 2285 |
+
"eval_loss": 16.56247329711914,
|
| 2286 |
+
"eval_runtime": 48.8722,
|
| 2287 |
+
"eval_samples_per_second": 386.886,
|
| 2288 |
+
"eval_steps_per_second": 1.514,
|
| 2289 |
+
"step": 71000
|
| 2290 |
+
},
|
| 2291 |
+
{
|
| 2292 |
+
"epoch": 1.751764014112113,
|
| 2293 |
+
"grad_norm": 11.369614601135254,
|
| 2294 |
+
"learning_rate": 2.559523435232101e-05,
|
| 2295 |
+
"loss": 0.4287,
|
| 2296 |
+
"step": 71500
|
| 2297 |
+
},
|
| 2298 |
+
{
|
| 2299 |
+
"epoch": 1.751764014112113,
|
| 2300 |
+
"eval_cosine_accuracy": 0.7656894326210022,
|
| 2301 |
+
"eval_loss": 16.205015182495117,
|
| 2302 |
+
"eval_runtime": 49.6805,
|
| 2303 |
+
"eval_samples_per_second": 380.592,
|
| 2304 |
+
"eval_steps_per_second": 1.49,
|
| 2305 |
+
"step": 71500
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 1.764014112112897,
|
| 2309 |
+
"grad_norm": 0.0,
|
| 2310 |
+
"learning_rate": 2.5458157380484434e-05,
|
| 2311 |
+
"loss": 0.4473,
|
| 2312 |
+
"step": 72000
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 1.764014112112897,
|
| 2316 |
+
"eval_cosine_accuracy": 0.7568535208702087,
|
| 2317 |
+
"eval_loss": 16.384424209594727,
|
| 2318 |
+
"eval_runtime": 49.5982,
|
| 2319 |
+
"eval_samples_per_second": 381.224,
|
| 2320 |
+
"eval_steps_per_second": 1.492,
|
| 2321 |
+
"step": 72000
|
| 2322 |
+
},
|
| 2323 |
+
{
|
| 2324 |
+
"epoch": 1.776264210113681,
|
| 2325 |
+
"grad_norm": 6.489040374755859,
|
| 2326 |
+
"learning_rate": 2.531935941809654e-05,
|
| 2327 |
+
"loss": 0.4481,
|
| 2328 |
+
"step": 72500
|
| 2329 |
+
},
|
| 2330 |
+
{
|
| 2331 |
+
"epoch": 1.776264210113681,
|
| 2332 |
+
"eval_cosine_accuracy": 0.7388191819190979,
|
| 2333 |
+
"eval_loss": 16.72735595703125,
|
| 2334 |
+
"eval_runtime": 49.6465,
|
| 2335 |
+
"eval_samples_per_second": 380.852,
|
| 2336 |
+
"eval_steps_per_second": 1.491,
|
| 2337 |
+
"step": 72500
|
| 2338 |
+
},
|
| 2339 |
+
{
|
| 2340 |
+
"epoch": 1.7885143081144648,
|
| 2341 |
+
"grad_norm": 6.377436637878418,
|
| 2342 |
+
"learning_rate": 2.5178863305699235e-05,
|
| 2343 |
+
"loss": 0.3861,
|
| 2344 |
+
"step": 73000
|
| 2345 |
+
},
|
| 2346 |
+
{
|
| 2347 |
+
"epoch": 1.7885143081144648,
|
| 2348 |
+
"eval_cosine_accuracy": 0.7689972519874573,
|
| 2349 |
+
"eval_loss": 15.925857543945312,
|
| 2350 |
+
"eval_runtime": 49.995,
|
| 2351 |
+
"eval_samples_per_second": 378.198,
|
| 2352 |
+
"eval_steps_per_second": 1.48,
|
| 2353 |
+
"step": 73000
|
| 2354 |
+
},
|
| 2355 |
+
{
|
| 2356 |
+
"epoch": 1.800764406115249,
|
| 2357 |
+
"grad_norm": 4.373341083526611,
|
| 2358 |
+
"learning_rate": 2.5036692163281347e-05,
|
| 2359 |
+
"loss": 0.4126,
|
| 2360 |
+
"step": 73500
|
| 2361 |
+
},
|
| 2362 |
+
{
|
| 2363 |
+
"epoch": 1.800764406115249,
|
| 2364 |
+
"eval_cosine_accuracy": 0.7714440822601318,
|
| 2365 |
+
"eval_loss": 16.406005859375,
|
| 2366 |
+
"eval_runtime": 49.0498,
|
| 2367 |
+
"eval_samples_per_second": 385.486,
|
| 2368 |
+
"eval_steps_per_second": 1.509,
|
| 2369 |
+
"step": 73500
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 1.8130145041160328,
|
| 2373 |
+
"grad_norm": 0.0,
|
| 2374 |
+
"learning_rate": 2.489315866467268e-05,
|
| 2375 |
+
"loss": 0.4328,
|
| 2376 |
+
"step": 74000
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 1.8130145041160328,
|
| 2380 |
+
"eval_cosine_accuracy": 0.7695409655570984,
|
| 2381 |
+
"eval_loss": 16.443193435668945,
|
| 2382 |
+
"eval_runtime": 48.8823,
|
| 2383 |
+
"eval_samples_per_second": 386.807,
|
| 2384 |
+
"eval_steps_per_second": 1.514,
|
| 2385 |
+
"step": 74000
|
| 2386 |
+
},
|
| 2387 |
+
{
|
| 2388 |
+
"epoch": 1.825264602116817,
|
| 2389 |
+
"grad_norm": 8.511246681213379,
|
| 2390 |
+
"learning_rate": 2.474771115303462e-05,
|
| 2391 |
+
"loss": 0.3848,
|
| 2392 |
+
"step": 74500
|
| 2393 |
+
},
|
| 2394 |
+
{
|
| 2395 |
+
"epoch": 1.825264602116817,
|
| 2396 |
+
"eval_cosine_accuracy": 0.7670941352844238,
|
| 2397 |
+
"eval_loss": 16.270015716552734,
|
| 2398 |
+
"eval_runtime": 50.4223,
|
| 2399 |
+
"eval_samples_per_second": 374.993,
|
| 2400 |
+
"eval_steps_per_second": 1.468,
|
| 2401 |
+
"step": 74500
|
| 2402 |
+
},
|
| 2403 |
+
{
|
| 2404 |
+
"epoch": 1.8375147001176009,
|
| 2405 |
+
"grad_norm": 5.887655735015869,
|
| 2406 |
+
"learning_rate": 2.460095524963032e-05,
|
| 2407 |
+
"loss": 0.4102,
|
| 2408 |
+
"step": 75000
|
| 2409 |
+
},
|
| 2410 |
+
{
|
| 2411 |
+
"epoch": 1.8375147001176009,
|
| 2412 |
+
"eval_cosine_accuracy": 0.7816393971443176,
|
| 2413 |
+
"eval_loss": 15.971381187438965,
|
| 2414 |
+
"eval_runtime": 50.7398,
|
| 2415 |
+
"eval_samples_per_second": 372.646,
|
| 2416 |
+
"eval_steps_per_second": 1.458,
|
| 2417 |
+
"step": 75000
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 1.849764798118385,
|
| 2421 |
+
"grad_norm": 10.97333812713623,
|
| 2422 |
+
"learning_rate": 2.4452326912626316e-05,
|
| 2423 |
+
"loss": 0.4078,
|
| 2424 |
+
"step": 75500
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 1.849764798118385,
|
| 2428 |
+
"eval_cosine_accuracy": 0.7499207258224487,
|
| 2429 |
+
"eval_loss": 16.55946922302246,
|
| 2430 |
+
"eval_runtime": 48.894,
|
| 2431 |
+
"eval_samples_per_second": 386.714,
|
| 2432 |
+
"eval_steps_per_second": 1.513,
|
| 2433 |
+
"step": 75500
|
| 2434 |
+
},
|
| 2435 |
+
{
|
| 2436 |
+
"epoch": 1.862014896119169,
|
| 2437 |
+
"grad_norm": 6.974571704864502,
|
| 2438 |
+
"learning_rate": 2.43021431041647e-05,
|
| 2439 |
+
"loss": 0.4184,
|
| 2440 |
+
"step": 76000
|
| 2441 |
+
},
|
| 2442 |
+
{
|
| 2443 |
+
"epoch": 1.862014896119169,
|
| 2444 |
+
"eval_cosine_accuracy": 0.752140998840332,
|
| 2445 |
+
"eval_loss": 16.734323501586914,
|
| 2446 |
+
"eval_runtime": 49.9348,
|
| 2447 |
+
"eval_samples_per_second": 378.653,
|
| 2448 |
+
"eval_steps_per_second": 1.482,
|
| 2449 |
+
"step": 76000
|
| 2450 |
+
},
|
| 2451 |
+
{
|
| 2452 |
+
"epoch": 1.874264994119953,
|
| 2453 |
+
"grad_norm": 0.0,
|
| 2454 |
+
"learning_rate": 2.4150428538438033e-05,
|
| 2455 |
+
"loss": 0.4054,
|
| 2456 |
+
"step": 76500
|
| 2457 |
+
},
|
| 2458 |
+
{
|
| 2459 |
+
"epoch": 1.874264994119953,
|
| 2460 |
+
"eval_cosine_accuracy": 0.7608410120010376,
|
| 2461 |
+
"eval_loss": 16.345609664916992,
|
| 2462 |
+
"eval_runtime": 48.8473,
|
| 2463 |
+
"eval_samples_per_second": 387.084,
|
| 2464 |
+
"eval_steps_per_second": 1.515,
|
| 2465 |
+
"step": 76500
|
| 2466 |
+
},
|
| 2467 |
+
{
|
| 2468 |
+
"epoch": 1.886515092120737,
|
| 2469 |
+
"grad_norm": 5.634564399719238,
|
| 2470 |
+
"learning_rate": 2.399720818153973e-05,
|
| 2471 |
+
"loss": 0.3862,
|
| 2472 |
+
"step": 77000
|
| 2473 |
+
},
|
| 2474 |
+
{
|
| 2475 |
+
"epoch": 1.886515092120737,
|
| 2476 |
+
"eval_cosine_accuracy": 0.7591191530227661,
|
| 2477 |
+
"eval_loss": 16.739871978759766,
|
| 2478 |
+
"eval_runtime": 49.7464,
|
| 2479 |
+
"eval_samples_per_second": 380.088,
|
| 2480 |
+
"eval_steps_per_second": 1.488,
|
| 2481 |
+
"step": 77000
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 1.898765190121521,
|
| 2485 |
+
"grad_norm": 8.459379196166992,
|
| 2486 |
+
"learning_rate": 2.384250724735564e-05,
|
| 2487 |
+
"loss": 0.4107,
|
| 2488 |
+
"step": 77500
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 1.898765190121521,
|
| 2492 |
+
"eval_cosine_accuracy": 0.7626988291740417,
|
| 2493 |
+
"eval_loss": 16.120464324951172,
|
| 2494 |
+
"eval_runtime": 49.4468,
|
| 2495 |
+
"eval_samples_per_second": 382.391,
|
| 2496 |
+
"eval_steps_per_second": 1.497,
|
| 2497 |
+
"step": 77500
|
| 2498 |
+
},
|
| 2499 |
+
{
|
| 2500 |
+
"epoch": 1.911015288122305,
|
| 2501 |
+
"grad_norm": 5.752081394195557,
|
| 2502 |
+
"learning_rate": 2.3686351193414895e-05,
|
| 2503 |
+
"loss": 0.3879,
|
| 2504 |
+
"step": 78000
|
| 2505 |
+
},
|
| 2506 |
+
{
|
| 2507 |
+
"epoch": 1.911015288122305,
|
| 2508 |
+
"eval_cosine_accuracy": 0.7634690999984741,
|
| 2509 |
+
"eval_loss": 16.300392150878906,
|
| 2510 |
+
"eval_runtime": 49.7421,
|
| 2511 |
+
"eval_samples_per_second": 380.121,
|
| 2512 |
+
"eval_steps_per_second": 1.488,
|
| 2513 |
+
"step": 78000
|
| 2514 |
+
},
|
| 2515 |
+
{
|
| 2516 |
+
"epoch": 1.9232653861230888,
|
| 2517 |
+
"grad_norm": 0.0,
|
| 2518 |
+
"learning_rate": 2.352876571670058e-05,
|
| 2519 |
+
"loss": 0.4032,
|
| 2520 |
+
"step": 78500
|
| 2521 |
+
},
|
| 2522 |
+
{
|
| 2523 |
+
"epoch": 1.9232653861230888,
|
| 2524 |
+
"eval_cosine_accuracy": 0.7540441155433655,
|
| 2525 |
+
"eval_loss": 16.190589904785156,
|
| 2526 |
+
"eval_runtime": 48.8496,
|
| 2527 |
+
"eval_samples_per_second": 387.066,
|
| 2528 |
+
"eval_steps_per_second": 1.515,
|
| 2529 |
+
"step": 78500
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 1.935515484123873,
|
| 2533 |
+
"grad_norm": 7.167166709899902,
|
| 2534 |
+
"learning_rate": 2.336977674942107e-05,
|
| 2535 |
+
"loss": 0.3979,
|
| 2536 |
+
"step": 79000
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 1.935515484123873,
|
| 2540 |
+
"eval_cosine_accuracy": 0.7497847676277161,
|
| 2541 |
+
"eval_loss": 16.35584831237793,
|
| 2542 |
+
"eval_runtime": 49.5472,
|
| 2543 |
+
"eval_samples_per_second": 381.616,
|
| 2544 |
+
"eval_steps_per_second": 1.494,
|
| 2545 |
+
"step": 79000
|
| 2546 |
+
},
|
| 2547 |
+
{
|
| 2548 |
+
"epoch": 1.947765582124657,
|
| 2549 |
+
"grad_norm": 5.026740074157715,
|
| 2550 |
+
"learning_rate": 2.3209732544401573e-05,
|
| 2551 |
+
"loss": 0.3721,
|
| 2552 |
+
"step": 79500
|
| 2553 |
+
},
|
| 2554 |
+
{
|
| 2555 |
+
"epoch": 1.947765582124657,
|
| 2556 |
+
"eval_cosine_accuracy": 0.7463863492012024,
|
| 2557 |
+
"eval_loss": 16.15901756286621,
|
| 2558 |
+
"eval_runtime": 48.8795,
|
| 2559 |
+
"eval_samples_per_second": 386.829,
|
| 2560 |
+
"eval_steps_per_second": 1.514,
|
| 2561 |
+
"step": 79500
|
| 2562 |
+
},
|
| 2563 |
+
{
|
| 2564 |
+
"epoch": 1.9600156801254411,
|
| 2565 |
+
"grad_norm": 9.04308795928955,
|
| 2566 |
+
"learning_rate": 2.3048017987532616e-05,
|
| 2567 |
+
"loss": 0.3895,
|
| 2568 |
+
"step": 80000
|
| 2569 |
+
},
|
| 2570 |
+
{
|
| 2571 |
+
"epoch": 1.9600156801254411,
|
| 2572 |
+
"eval_cosine_accuracy": 0.7588472366333008,
|
| 2573 |
+
"eval_loss": 16.208189010620117,
|
| 2574 |
+
"eval_runtime": 49.8276,
|
| 2575 |
+
"eval_samples_per_second": 379.469,
|
| 2576 |
+
"eval_steps_per_second": 1.485,
|
| 2577 |
+
"step": 80000
|
| 2578 |
+
},
|
| 2579 |
+
{
|
| 2580 |
+
"epoch": 1.972265778126225,
|
| 2581 |
+
"grad_norm": 6.253329753875732,
|
| 2582 |
+
"learning_rate": 2.2884979051768852e-05,
|
| 2583 |
+
"loss": 0.3997,
|
| 2584 |
+
"step": 80500
|
| 2585 |
+
},
|
| 2586 |
+
{
|
| 2587 |
+
"epoch": 1.972265778126225,
|
| 2588 |
+
"eval_cosine_accuracy": 0.7568082213401794,
|
| 2589 |
+
"eval_loss": 16.452190399169922,
|
| 2590 |
+
"eval_runtime": 48.7985,
|
| 2591 |
+
"eval_samples_per_second": 387.471,
|
| 2592 |
+
"eval_steps_per_second": 1.516,
|
| 2593 |
+
"step": 80500
|
| 2594 |
+
},
|
| 2595 |
+
{
|
| 2596 |
+
"epoch": 1.984515876127009,
|
| 2597 |
+
"grad_norm": 7.272450923919678,
|
| 2598 |
+
"learning_rate": 2.2720642566737875e-05,
|
| 2599 |
+
"loss": 0.3732,
|
| 2600 |
+
"step": 81000
|
| 2601 |
+
},
|
| 2602 |
+
{
|
| 2603 |
+
"epoch": 1.984515876127009,
|
| 2604 |
+
"eval_cosine_accuracy": 0.7486972808837891,
|
| 2605 |
+
"eval_loss": 16.443246841430664,
|
| 2606 |
+
"eval_runtime": 49.6729,
|
| 2607 |
+
"eval_samples_per_second": 380.651,
|
| 2608 |
+
"eval_steps_per_second": 1.49,
|
| 2609 |
+
"step": 81000
|
| 2610 |
+
},
|
| 2611 |
+
{
|
| 2612 |
+
"epoch": 1.996765974127793,
|
| 2613 |
+
"grad_norm": 0.0,
|
| 2614 |
+
"learning_rate": 2.2555035575591528e-05,
|
| 2615 |
+
"loss": 0.394,
|
| 2616 |
+
"step": 81500
|
| 2617 |
+
},
|
| 2618 |
+
{
|
| 2619 |
+
"epoch": 1.996765974127793,
|
| 2620 |
+
"eval_cosine_accuracy": 0.7561737895011902,
|
| 2621 |
+
"eval_loss": 16.488916397094727,
|
| 2622 |
+
"eval_runtime": 48.9116,
|
| 2623 |
+
"eval_samples_per_second": 386.575,
|
| 2624 |
+
"eval_steps_per_second": 1.513,
|
| 2625 |
+
"step": 81500
|
| 2626 |
+
}
|
| 2627 |
+
],
|
| 2628 |
+
"logging_steps": 500,
|
| 2629 |
+
"max_steps": 163264,
|
| 2630 |
+
"num_input_tokens_seen": 0,
|
| 2631 |
+
"num_train_epochs": 4,
|
| 2632 |
+
"save_steps": 500,
|
| 2633 |
+
"stateful_callbacks": {
|
| 2634 |
+
"TrainerControl": {
|
| 2635 |
+
"args": {
|
| 2636 |
+
"should_epoch_stop": false,
|
| 2637 |
+
"should_evaluate": false,
|
| 2638 |
+
"should_log": false,
|
| 2639 |
+
"should_save": true,
|
| 2640 |
+
"should_training_stop": false
|
| 2641 |
+
},
|
| 2642 |
+
"attributes": {}
|
| 2643 |
+
}
|
| 2644 |
+
},
|
| 2645 |
+
"total_flos": 0.0,
|
| 2646 |
+
"train_batch_size": 256,
|
| 2647 |
+
"trial_name": null,
|
| 2648 |
+
"trial_params": null
|
| 2649 |
+
}
|
checkpoint-81632/training_metrics.json
ADDED
|
@@ -0,0 +1,984 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metrics_history": [
|
| 3 |
+
{
|
| 4 |
+
"epoch": 0.012250098000784006,
|
| 5 |
+
"global_step": 500,
|
| 6 |
+
"eval_loss": 13.151524543762207,
|
| 7 |
+
"eval_cosine_accuracy": 0.9715437889099121
|
| 8 |
+
},
|
| 9 |
+
{
|
| 10 |
+
"epoch": 0.02450019600156801,
|
| 11 |
+
"global_step": 1000,
|
| 12 |
+
"eval_loss": 8.770788192749023,
|
| 13 |
+
"eval_cosine_accuracy": 0.9591735005378723
|
| 14 |
+
},
|
| 15 |
+
{
|
| 16 |
+
"epoch": 0.03675029400235202,
|
| 17 |
+
"global_step": 1500,
|
| 18 |
+
"eval_loss": 8.737192153930664,
|
| 19 |
+
"eval_cosine_accuracy": 0.9491594433784485
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"epoch": 0.04900039200313602,
|
| 23 |
+
"global_step": 2000,
|
| 24 |
+
"eval_loss": 9.47952651977539,
|
| 25 |
+
"eval_cosine_accuracy": 0.9433141350746155
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.06125049000392003,
|
| 29 |
+
"global_step": 2500,
|
| 30 |
+
"eval_loss": 9.841277122497559,
|
| 31 |
+
"eval_cosine_accuracy": 0.939236044883728
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.07350058800470403,
|
| 35 |
+
"global_step": 3000,
|
| 36 |
+
"eval_loss": 10.269918441772461,
|
| 37 |
+
"eval_cosine_accuracy": 0.9352032542228699
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.08575068600548805,
|
| 41 |
+
"global_step": 3500,
|
| 42 |
+
"eval_loss": 10.558481216430664,
|
| 43 |
+
"eval_cosine_accuracy": 0.9305813312530518
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 0.09800078400627205,
|
| 47 |
+
"global_step": 4000,
|
| 48 |
+
"eval_loss": 10.612824440002441,
|
| 49 |
+
"eval_cosine_accuracy": 0.9273641705513
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.11025088200705606,
|
| 53 |
+
"global_step": 4500,
|
| 54 |
+
"eval_loss": 11.195393562316895,
|
| 55 |
+
"eval_cosine_accuracy": 0.9201141595840454
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.12250098000784006,
|
| 59 |
+
"global_step": 5000,
|
| 60 |
+
"eval_loss": 11.262394905090332,
|
| 61 |
+
"eval_cosine_accuracy": 0.9150844812393188
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.13475107800862407,
|
| 65 |
+
"global_step": 5500,
|
| 66 |
+
"eval_loss": 11.321725845336914,
|
| 67 |
+
"eval_cosine_accuracy": 0.9161266684532166
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.14700117600940807,
|
| 71 |
+
"global_step": 6000,
|
| 72 |
+
"eval_loss": 11.53966236114502,
|
| 73 |
+
"eval_cosine_accuracy": 0.9153110980987549
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.1592512740101921,
|
| 77 |
+
"global_step": 6500,
|
| 78 |
+
"eval_loss": 11.740141868591309,
|
| 79 |
+
"eval_cosine_accuracy": 0.9076985716819763
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.1715013720109761,
|
| 83 |
+
"global_step": 7000,
|
| 84 |
+
"eval_loss": 11.773887634277344,
|
| 85 |
+
"eval_cosine_accuracy": 0.9086048007011414
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.1837514700117601,
|
| 89 |
+
"global_step": 7500,
|
| 90 |
+
"eval_loss": 12.162515640258789,
|
| 91 |
+
"eval_cosine_accuracy": 0.9028048515319824
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.1960015680125441,
|
| 95 |
+
"global_step": 8000,
|
| 96 |
+
"eval_loss": 11.9808349609375,
|
| 97 |
+
"eval_cosine_accuracy": 0.9039376378059387
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.20825166601332812,
|
| 101 |
+
"global_step": 8500,
|
| 102 |
+
"eval_loss": 12.287437438964844,
|
| 103 |
+
"eval_cosine_accuracy": 0.9011735916137695
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 0.22050176401411212,
|
| 107 |
+
"global_step": 9000,
|
| 108 |
+
"eval_loss": 12.496896743774414,
|
| 109 |
+
"eval_cosine_accuracy": 0.8947392106056213
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.23275186201489612,
|
| 113 |
+
"global_step": 9500,
|
| 114 |
+
"eval_loss": 12.531494140625,
|
| 115 |
+
"eval_cosine_accuracy": 0.8911142349243164
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.24500196001568011,
|
| 119 |
+
"global_step": 10000,
|
| 120 |
+
"eval_loss": 12.771232604980469,
|
| 121 |
+
"eval_cosine_accuracy": 0.8850423693656921
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.2572520580164641,
|
| 125 |
+
"global_step": 10500,
|
| 126 |
+
"eval_loss": 12.552855491638184,
|
| 127 |
+
"eval_cosine_accuracy": 0.8918845653533936
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 0.26950215601724814,
|
| 131 |
+
"global_step": 11000,
|
| 132 |
+
"eval_loss": 12.82436752319336,
|
| 133 |
+
"eval_cosine_accuracy": 0.885540783405304
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 0.28175225401803217,
|
| 137 |
+
"global_step": 11500,
|
| 138 |
+
"eval_loss": 12.749467849731445,
|
| 139 |
+
"eval_cosine_accuracy": 0.8915220499038696
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.29400235201881614,
|
| 143 |
+
"global_step": 12000,
|
| 144 |
+
"eval_loss": 12.727092742919922,
|
| 145 |
+
"eval_cosine_accuracy": 0.8872627019882202
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.30625245001960016,
|
| 149 |
+
"global_step": 12500,
|
| 150 |
+
"eval_loss": 12.634977340698242,
|
| 151 |
+
"eval_cosine_accuracy": 0.8861298561096191
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.3185025480203842,
|
| 155 |
+
"global_step": 13000,
|
| 156 |
+
"eval_loss": 12.859210014343262,
|
| 157 |
+
"eval_cosine_accuracy": 0.8913407921791077
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.33075264602116816,
|
| 161 |
+
"global_step": 13500,
|
| 162 |
+
"eval_loss": 12.690762519836426,
|
| 163 |
+
"eval_cosine_accuracy": 0.8802845478057861
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.3430027440219522,
|
| 167 |
+
"global_step": 14000,
|
| 168 |
+
"eval_loss": 13.312505722045898,
|
| 169 |
+
"eval_cosine_accuracy": 0.87131267786026
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.35525284202273616,
|
| 173 |
+
"global_step": 14500,
|
| 174 |
+
"eval_loss": 13.471542358398438,
|
| 175 |
+
"eval_cosine_accuracy": 0.8672345876693726
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 0.3675029400235202,
|
| 179 |
+
"global_step": 15000,
|
| 180 |
+
"eval_loss": 12.901288032531738,
|
| 181 |
+
"eval_cosine_accuracy": 0.8804205060005188
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.3797530380243042,
|
| 185 |
+
"global_step": 15500,
|
| 186 |
+
"eval_loss": 13.632067680358887,
|
| 187 |
+
"eval_cosine_accuracy": 0.861298680305481
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 0.3920031360250882,
|
| 191 |
+
"global_step": 16000,
|
| 192 |
+
"eval_loss": 12.429121017456055,
|
| 193 |
+
"eval_cosine_accuracy": 0.876070499420166
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.4042532340258722,
|
| 197 |
+
"global_step": 16500,
|
| 198 |
+
"eval_loss": 13.132347106933594,
|
| 199 |
+
"eval_cosine_accuracy": 0.8699533343315125
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.41650333202665624,
|
| 203 |
+
"global_step": 17000,
|
| 204 |
+
"eval_loss": 13.065674781799316,
|
| 205 |
+
"eval_cosine_accuracy": 0.8669173717498779
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.4287534300274402,
|
| 209 |
+
"global_step": 17500,
|
| 210 |
+
"eval_loss": 13.467134475708008,
|
| 211 |
+
"eval_cosine_accuracy": 0.8608002066612244
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.44100352802822423,
|
| 215 |
+
"global_step": 18000,
|
| 216 |
+
"eval_loss": 13.483769416809082,
|
| 217 |
+
"eval_cosine_accuracy": 0.8644705414772034
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.4532536260290082,
|
| 221 |
+
"global_step": 18500,
|
| 222 |
+
"eval_loss": 13.853730201721191,
|
| 223 |
+
"eval_cosine_accuracy": 0.8525533676147461
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 0.46550372402979223,
|
| 227 |
+
"global_step": 19000,
|
| 228 |
+
"eval_loss": 13.518957138061523,
|
| 229 |
+
"eval_cosine_accuracy": 0.8601658344268799
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 0.47775382203057626,
|
| 233 |
+
"global_step": 19500,
|
| 234 |
+
"eval_loss": 14.119278907775879,
|
| 235 |
+
"eval_cosine_accuracy": 0.8497439622879028
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 0.49000392003136023,
|
| 239 |
+
"global_step": 20000,
|
| 240 |
+
"eval_loss": 13.421476364135742,
|
| 241 |
+
"eval_cosine_accuracy": 0.8596221208572388
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.5022540180321443,
|
| 245 |
+
"global_step": 20500,
|
| 246 |
+
"eval_loss": 13.997146606445312,
|
| 247 |
+
"eval_cosine_accuracy": 0.8446236848831177
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.5145041160329282,
|
| 251 |
+
"global_step": 21000,
|
| 252 |
+
"eval_loss": 13.374435424804688,
|
| 253 |
+
"eval_cosine_accuracy": 0.863473653793335
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.5267542140337123,
|
| 257 |
+
"global_step": 21500,
|
| 258 |
+
"eval_loss": 13.285289764404297,
|
| 259 |
+
"eval_cosine_accuracy": 0.857628345489502
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 0.5390043120344963,
|
| 263 |
+
"global_step": 22000,
|
| 264 |
+
"eval_loss": 13.88798713684082,
|
| 265 |
+
"eval_cosine_accuracy": 0.8546377420425415
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.5512544100352803,
|
| 269 |
+
"global_step": 22500,
|
| 270 |
+
"eval_loss": 13.286218643188477,
|
| 271 |
+
"eval_cosine_accuracy": 0.8662376999855042
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.5635045080360643,
|
| 275 |
+
"global_step": 23000,
|
| 276 |
+
"eval_loss": 14.143769264221191,
|
| 277 |
+
"eval_cosine_accuracy": 0.8439440131187439
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 0.5757546060368482,
|
| 281 |
+
"global_step": 23500,
|
| 282 |
+
"eval_loss": 14.088532447814941,
|
| 283 |
+
"eval_cosine_accuracy": 0.8401830792427063
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.5880047040376323,
|
| 287 |
+
"global_step": 24000,
|
| 288 |
+
"eval_loss": 13.797255516052246,
|
| 289 |
+
"eval_cosine_accuracy": 0.848339319229126
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.6002548020384163,
|
| 293 |
+
"global_step": 24500,
|
| 294 |
+
"eval_loss": 13.715460777282715,
|
| 295 |
+
"eval_cosine_accuracy": 0.8492002487182617
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.6125049000392003,
|
| 299 |
+
"global_step": 25000,
|
| 300 |
+
"eval_loss": 14.3268461227417,
|
| 301 |
+
"eval_cosine_accuracy": 0.8482486605644226
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 0.6247549980399844,
|
| 305 |
+
"global_step": 25500,
|
| 306 |
+
"eval_loss": 13.635766983032227,
|
| 307 |
+
"eval_cosine_accuracy": 0.8472517728805542
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 0.6370050960407684,
|
| 311 |
+
"global_step": 26000,
|
| 312 |
+
"eval_loss": 14.338939666748047,
|
| 313 |
+
"eval_cosine_accuracy": 0.835198700428009
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.6492551940415523,
|
| 317 |
+
"global_step": 26500,
|
| 318 |
+
"eval_loss": 14.60715103149414,
|
| 319 |
+
"eval_cosine_accuracy": 0.8212424516677856
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 0.6615052920423363,
|
| 323 |
+
"global_step": 27000,
|
| 324 |
+
"eval_loss": 14.688072204589844,
|
| 325 |
+
"eval_cosine_accuracy": 0.8262268304824829
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.6737553900431204,
|
| 329 |
+
"global_step": 27500,
|
| 330 |
+
"eval_loss": 14.224186897277832,
|
| 331 |
+
"eval_cosine_accuracy": 0.8329330682754517
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.6860054880439044,
|
| 335 |
+
"global_step": 28000,
|
| 336 |
+
"eval_loss": 14.43342399597168,
|
| 337 |
+
"eval_cosine_accuracy": 0.8453940153121948
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 0.6982555860446884,
|
| 341 |
+
"global_step": 28500,
|
| 342 |
+
"eval_loss": 13.982333183288574,
|
| 343 |
+
"eval_cosine_accuracy": 0.8401377201080322
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 0.7105056840454723,
|
| 347 |
+
"global_step": 29000,
|
| 348 |
+
"eval_loss": 14.23220443725586,
|
| 349 |
+
"eval_cosine_accuracy": 0.8279939889907837
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 0.7227557820462563,
|
| 353 |
+
"global_step": 29500,
|
| 354 |
+
"eval_loss": 14.843314170837402,
|
| 355 |
+
"eval_cosine_accuracy": 0.8259549736976624
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 0.7350058800470404,
|
| 359 |
+
"global_step": 30000,
|
| 360 |
+
"eval_loss": 14.670516014099121,
|
| 361 |
+
"eval_cosine_accuracy": 0.8178893327713013
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.7472559780478244,
|
| 365 |
+
"global_step": 30500,
|
| 366 |
+
"eval_loss": 14.975008964538574,
|
| 367 |
+
"eval_cosine_accuracy": 0.813086211681366
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.7595060760486084,
|
| 371 |
+
"global_step": 31000,
|
| 372 |
+
"eval_loss": 14.855353355407715,
|
| 373 |
+
"eval_cosine_accuracy": 0.8202909231185913
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.7717561740493923,
|
| 377 |
+
"global_step": 31500,
|
| 378 |
+
"eval_loss": 14.634927749633789,
|
| 379 |
+
"eval_cosine_accuracy": 0.8221487402915955
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 0.7840062720501764,
|
| 383 |
+
"global_step": 32000,
|
| 384 |
+
"eval_loss": 15.120546340942383,
|
| 385 |
+
"eval_cosine_accuracy": 0.8112737536430359
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 0.7962563700509604,
|
| 389 |
+
"global_step": 32500,
|
| 390 |
+
"eval_loss": 14.621186256408691,
|
| 391 |
+
"eval_cosine_accuracy": 0.8139471411705017
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.8085064680517444,
|
| 395 |
+
"global_step": 33000,
|
| 396 |
+
"eval_loss": 14.621684074401855,
|
| 397 |
+
"eval_cosine_accuracy": 0.8144456148147583
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 0.8207565660525284,
|
| 401 |
+
"global_step": 33500,
|
| 402 |
+
"eval_loss": 15.029080390930176,
|
| 403 |
+
"eval_cosine_accuracy": 0.8181612491607666
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.8330066640533125,
|
| 407 |
+
"global_step": 34000,
|
| 408 |
+
"eval_loss": 14.72106647491455,
|
| 409 |
+
"eval_cosine_accuracy": 0.8164846897125244
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.8452567620540964,
|
| 413 |
+
"global_step": 34500,
|
| 414 |
+
"eval_loss": 14.777338981628418,
|
| 415 |
+
"eval_cosine_accuracy": 0.8250033855438232
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.8575068600548804,
|
| 419 |
+
"global_step": 35000,
|
| 420 |
+
"eval_loss": 13.815231323242188,
|
| 421 |
+
"eval_cosine_accuracy": 0.8253206014633179
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 0.8697569580556644,
|
| 425 |
+
"global_step": 35500,
|
| 426 |
+
"eval_loss": 14.613081932067871,
|
| 427 |
+
"eval_cosine_accuracy": 0.8259549736976624
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 0.8820070560564485,
|
| 431 |
+
"global_step": 36000,
|
| 432 |
+
"eval_loss": 14.53075885772705,
|
| 433 |
+
"eval_cosine_accuracy": 0.8182971477508545
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.8942571540572325,
|
| 437 |
+
"global_step": 36500,
|
| 438 |
+
"eval_loss": 14.517987251281738,
|
| 439 |
+
"eval_cosine_accuracy": 0.8179346323013306
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 0.9065072520580164,
|
| 443 |
+
"global_step": 37000,
|
| 444 |
+
"eval_loss": 14.480570793151855,
|
| 445 |
+
"eval_cosine_accuracy": 0.8172096610069275
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 0.9187573500588004,
|
| 449 |
+
"global_step": 37500,
|
| 450 |
+
"eval_loss": 15.179258346557617,
|
| 451 |
+
"eval_cosine_accuracy": 0.8054737448692322
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.9310074480595845,
|
| 455 |
+
"global_step": 38000,
|
| 456 |
+
"eval_loss": 15.210185050964355,
|
| 457 |
+
"eval_cosine_accuracy": 0.8111377954483032
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.9432575460603685,
|
| 461 |
+
"global_step": 38500,
|
| 462 |
+
"eval_loss": 14.83116340637207,
|
| 463 |
+
"eval_cosine_accuracy": 0.8057456016540527
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 0.9555076440611525,
|
| 467 |
+
"global_step": 39000,
|
| 468 |
+
"eval_loss": 15.22596549987793,
|
| 469 |
+
"eval_cosine_accuracy": 0.8018487691879272
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 0.9677577420619365,
|
| 473 |
+
"global_step": 39500,
|
| 474 |
+
"eval_loss": 15.166121482849121,
|
| 475 |
+
"eval_cosine_accuracy": 0.8076034188270569
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 0.9800078400627205,
|
| 479 |
+
"global_step": 40000,
|
| 480 |
+
"eval_loss": 15.132609367370605,
|
| 481 |
+
"eval_cosine_accuracy": 0.7957315444946289
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 0.9922579380635045,
|
| 485 |
+
"global_step": 40500,
|
| 486 |
+
"eval_loss": 15.568105697631836,
|
| 487 |
+
"eval_cosine_accuracy": 0.7949612736701965
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 1.0045080360642886,
|
| 491 |
+
"global_step": 41000,
|
| 492 |
+
"eval_loss": 15.345169067382812,
|
| 493 |
+
"eval_cosine_accuracy": 0.7948253154754639
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 1.0167581340650724,
|
| 497 |
+
"global_step": 41500,
|
| 498 |
+
"eval_loss": 15.186988830566406,
|
| 499 |
+
"eval_cosine_accuracy": 0.7893425226211548
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.0290082320658565,
|
| 503 |
+
"global_step": 42000,
|
| 504 |
+
"eval_loss": 14.84115982055664,
|
| 505 |
+
"eval_cosine_accuracy": 0.8046128153800964
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 1.0412583300666405,
|
| 509 |
+
"global_step": 42500,
|
| 510 |
+
"eval_loss": 15.067384719848633,
|
| 511 |
+
"eval_cosine_accuracy": 0.8026190400123596
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 1.0535084280674245,
|
| 515 |
+
"global_step": 43000,
|
| 516 |
+
"eval_loss": 14.875770568847656,
|
| 517 |
+
"eval_cosine_accuracy": 0.8100050091743469
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 1.0657585260682085,
|
| 521 |
+
"global_step": 43500,
|
| 522 |
+
"eval_loss": 15.206002235412598,
|
| 523 |
+
"eval_cosine_accuracy": 0.7897050380706787
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 1.0780086240689926,
|
| 527 |
+
"global_step": 44000,
|
| 528 |
+
"eval_loss": 14.952627182006836,
|
| 529 |
+
"eval_cosine_accuracy": 0.7938737869262695
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 1.0902587220697766,
|
| 533 |
+
"global_step": 44500,
|
| 534 |
+
"eval_loss": 16.021696090698242,
|
| 535 |
+
"eval_cosine_accuracy": 0.7745253443717957
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 1.1025088200705606,
|
| 539 |
+
"global_step": 45000,
|
| 540 |
+
"eval_loss": 15.794776916503906,
|
| 541 |
+
"eval_cosine_accuracy": 0.7867597341537476
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.1147589180713446,
|
| 545 |
+
"global_step": 45500,
|
| 546 |
+
"eval_loss": 15.574027061462402,
|
| 547 |
+
"eval_cosine_accuracy": 0.7909737825393677
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 1.1270090160721287,
|
| 551 |
+
"global_step": 46000,
|
| 552 |
+
"eval_loss": 15.986827850341797,
|
| 553 |
+
"eval_cosine_accuracy": 0.7768362760543823
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 1.1392591140729125,
|
| 557 |
+
"global_step": 46500,
|
| 558 |
+
"eval_loss": 16.077783584594727,
|
| 559 |
+
"eval_cosine_accuracy": 0.7769269347190857
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 1.1515092120736965,
|
| 563 |
+
"global_step": 47000,
|
| 564 |
+
"eval_loss": 15.513272285461426,
|
| 565 |
+
"eval_cosine_accuracy": 0.7923331260681152
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 1.1637593100744805,
|
| 569 |
+
"global_step": 47500,
|
| 570 |
+
"eval_loss": 15.18636703491211,
|
| 571 |
+
"eval_cosine_accuracy": 0.8087815642356873
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 1.1760094080752646,
|
| 575 |
+
"global_step": 48000,
|
| 576 |
+
"eval_loss": 15.985566139221191,
|
| 577 |
+
"eval_cosine_accuracy": 0.7749331593513489
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 1.1882595060760486,
|
| 581 |
+
"global_step": 48500,
|
| 582 |
+
"eval_loss": 15.625438690185547,
|
| 583 |
+
"eval_cosine_accuracy": 0.7782862782478333
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.2005096040768326,
|
| 587 |
+
"global_step": 49000,
|
| 588 |
+
"eval_loss": 15.482627868652344,
|
| 589 |
+
"eval_cosine_accuracy": 0.7873034477233887
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 1.2127597020776166,
|
| 593 |
+
"global_step": 49500,
|
| 594 |
+
"eval_loss": 15.559972763061523,
|
| 595 |
+
"eval_cosine_accuracy": 0.7844034433364868
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 1.2250098000784007,
|
| 599 |
+
"global_step": 50000,
|
| 600 |
+
"eval_loss": 15.543310165405273,
|
| 601 |
+
"eval_cosine_accuracy": 0.7828628420829773
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 1.2372598980791847,
|
| 605 |
+
"global_step": 50500,
|
| 606 |
+
"eval_loss": 15.88884162902832,
|
| 607 |
+
"eval_cosine_accuracy": 0.7695409655570984
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 1.2495099960799687,
|
| 611 |
+
"global_step": 51000,
|
| 612 |
+
"eval_loss": 15.716775894165039,
|
| 613 |
+
"eval_cosine_accuracy": 0.7802800536155701
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 1.2617600940807527,
|
| 617 |
+
"global_step": 51500,
|
| 618 |
+
"eval_loss": 15.803173065185547,
|
| 619 |
+
"eval_cosine_accuracy": 0.7801893949508667
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 1.2740101920815365,
|
| 623 |
+
"global_step": 52000,
|
| 624 |
+
"eval_loss": 16.11168670654297,
|
| 625 |
+
"eval_cosine_accuracy": 0.7831346988677979
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.2862602900823208,
|
| 629 |
+
"global_step": 52500,
|
| 630 |
+
"eval_loss": 16.117074966430664,
|
| 631 |
+
"eval_cosine_accuracy": 0.7808237671852112
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 1.2985103880831046,
|
| 635 |
+
"global_step": 53000,
|
| 636 |
+
"eval_loss": 15.681249618530273,
|
| 637 |
+
"eval_cosine_accuracy": 0.7743440866470337
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 1.3107604860838886,
|
| 641 |
+
"global_step": 53500,
|
| 642 |
+
"eval_loss": 15.613662719726562,
|
| 643 |
+
"eval_cosine_accuracy": 0.764194130897522
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 1.3230105840846726,
|
| 647 |
+
"global_step": 54000,
|
| 648 |
+
"eval_loss": 15.998211860656738,
|
| 649 |
+
"eval_cosine_accuracy": 0.7777425050735474
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 1.3352606820854567,
|
| 653 |
+
"global_step": 54500,
|
| 654 |
+
"eval_loss": 16.415210723876953,
|
| 655 |
+
"eval_cosine_accuracy": 0.7656894326210022
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 1.3475107800862407,
|
| 659 |
+
"global_step": 55000,
|
| 660 |
+
"eval_loss": 15.649203300476074,
|
| 661 |
+
"eval_cosine_accuracy": 0.766187846660614
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.3597608780870247,
|
| 665 |
+
"global_step": 55500,
|
| 666 |
+
"eval_loss": 16.437679290771484,
|
| 667 |
+
"eval_cosine_accuracy": 0.7639675736427307
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.3720109760878088,
|
| 671 |
+
"global_step": 56000,
|
| 672 |
+
"eval_loss": 16.5505313873291,
|
| 673 |
+
"eval_cosine_accuracy": 0.7617019414901733
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 1.3842610740885928,
|
| 677 |
+
"global_step": 56500,
|
| 678 |
+
"eval_loss": 16.079374313354492,
|
| 679 |
+
"eval_cosine_accuracy": 0.7699940800666809
|
| 680 |
+
},
|
| 681 |
+
{
|
| 682 |
+
"epoch": 1.3965111720893768,
|
| 683 |
+
"global_step": 57000,
|
| 684 |
+
"eval_loss": 15.54992961883545,
|
| 685 |
+
"eval_cosine_accuracy": 0.7776972055435181
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 1.4087612700901606,
|
| 689 |
+
"global_step": 57500,
|
| 690 |
+
"eval_loss": 15.954768180847168,
|
| 691 |
+
"eval_cosine_accuracy": 0.7788300514221191
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 1.4210113680909449,
|
| 695 |
+
"global_step": 58000,
|
| 696 |
+
"eval_loss": 16.53635597229004,
|
| 697 |
+
"eval_cosine_accuracy": 0.7652363181114197
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 1.4332614660917287,
|
| 701 |
+
"global_step": 58500,
|
| 702 |
+
"eval_loss": 15.921165466308594,
|
| 703 |
+
"eval_cosine_accuracy": 0.7767003774642944
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 1.4455115640925127,
|
| 707 |
+
"global_step": 59000,
|
| 708 |
+
"eval_loss": 16.090425491333008,
|
| 709 |
+
"eval_cosine_accuracy": 0.7660519480705261
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.4577616620932967,
|
| 713 |
+
"global_step": 59500,
|
| 714 |
+
"eval_loss": 16.565155029296875,
|
| 715 |
+
"eval_cosine_accuracy": 0.7597535252571106
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 1.4700117600940807,
|
| 719 |
+
"global_step": 60000,
|
| 720 |
+
"eval_loss": 16.681304931640625,
|
| 721 |
+
"eval_cosine_accuracy": 0.7472019791603088
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 1.4822618580948648,
|
| 725 |
+
"global_step": 60500,
|
| 726 |
+
"eval_loss": 16.14435577392578,
|
| 727 |
+
"eval_cosine_accuracy": 0.7678191065788269
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 1.4945119560956488,
|
| 731 |
+
"global_step": 61000,
|
| 732 |
+
"eval_loss": 16.488277435302734,
|
| 733 |
+
"eval_cosine_accuracy": 0.7491050958633423
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 1.5067620540964328,
|
| 737 |
+
"global_step": 61500,
|
| 738 |
+
"eval_loss": 16.454145431518555,
|
| 739 |
+
"eval_cosine_accuracy": 0.7501019239425659
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 1.5190121520972166,
|
| 743 |
+
"global_step": 62000,
|
| 744 |
+
"eval_loss": 16.385787963867188,
|
| 745 |
+
"eval_cosine_accuracy": 0.751053512096405
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 1.5312622500980009,
|
| 749 |
+
"global_step": 62500,
|
| 750 |
+
"eval_loss": 16.61661720275879,
|
| 751 |
+
"eval_cosine_accuracy": 0.7569894194602966
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.5435123480987847,
|
| 755 |
+
"global_step": 63000,
|
| 756 |
+
"eval_loss": 16.342727661132812,
|
| 757 |
+
"eval_cosine_accuracy": 0.7584394216537476
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 1.555762446099569,
|
| 761 |
+
"global_step": 63500,
|
| 762 |
+
"eval_loss": 16.419708251953125,
|
| 763 |
+
"eval_cosine_accuracy": 0.7613394260406494
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 1.5680125441003527,
|
| 767 |
+
"global_step": 64000,
|
| 768 |
+
"eval_loss": 15.770021438598633,
|
| 769 |
+
"eval_cosine_accuracy": 0.7609316110610962
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 1.5802626421011368,
|
| 773 |
+
"global_step": 64500,
|
| 774 |
+
"eval_loss": 16.45221519470215,
|
| 775 |
+
"eval_cosine_accuracy": 0.752503514289856
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 1.5925127401019208,
|
| 779 |
+
"global_step": 65000,
|
| 780 |
+
"eval_loss": 16.404518127441406,
|
| 781 |
+
"eval_cosine_accuracy": 0.7553128600120544
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 1.6047628381027048,
|
| 785 |
+
"global_step": 65500,
|
| 786 |
+
"eval_loss": 16.57503890991211,
|
| 787 |
+
"eval_cosine_accuracy": 0.7507363557815552
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 1.6170129361034888,
|
| 791 |
+
"global_step": 66000,
|
| 792 |
+
"eval_loss": 16.23915672302246,
|
| 793 |
+
"eval_cosine_accuracy": 0.7636956572532654
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.6292630341042729,
|
| 797 |
+
"global_step": 66500,
|
| 798 |
+
"eval_loss": 16.42445945739746,
|
| 799 |
+
"eval_cosine_accuracy": 0.7583941221237183
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 1.641513132105057,
|
| 803 |
+
"global_step": 67000,
|
| 804 |
+
"eval_loss": 15.90536880493164,
|
| 805 |
+
"eval_cosine_accuracy": 0.7737550139427185
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 1.6537632301058407,
|
| 809 |
+
"global_step": 67500,
|
| 810 |
+
"eval_loss": 16.559391021728516,
|
| 811 |
+
"eval_cosine_accuracy": 0.7579410076141357
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 1.666013328106625,
|
| 815 |
+
"global_step": 68000,
|
| 816 |
+
"eval_loss": 16.645488739013672,
|
| 817 |
+
"eval_cosine_accuracy": 0.763106644153595
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 1.6782634261074088,
|
| 821 |
+
"global_step": 68500,
|
| 822 |
+
"eval_loss": 16.888395309448242,
|
| 823 |
+
"eval_cosine_accuracy": 0.7497847676277161
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 1.690513524108193,
|
| 827 |
+
"global_step": 69000,
|
| 828 |
+
"eval_loss": 16.116737365722656,
|
| 829 |
+
"eval_cosine_accuracy": 0.7763378620147705
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 1.7027636221089768,
|
| 833 |
+
"global_step": 69500,
|
| 834 |
+
"eval_loss": 16.68659019470215,
|
| 835 |
+
"eval_cosine_accuracy": 0.7625628709793091
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 1.7150137201097608,
|
| 839 |
+
"global_step": 70000,
|
| 840 |
+
"eval_loss": 16.457317352294922,
|
| 841 |
+
"eval_cosine_accuracy": 0.7486066222190857
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 1.7272638181105449,
|
| 845 |
+
"global_step": 70500,
|
| 846 |
+
"eval_loss": 16.359926223754883,
|
| 847 |
+
"eval_cosine_accuracy": 0.7486066222190857
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 1.7395139161113289,
|
| 851 |
+
"global_step": 71000,
|
| 852 |
+
"eval_loss": 16.56247329711914,
|
| 853 |
+
"eval_cosine_accuracy": 0.7649644017219543
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 1.751764014112113,
|
| 857 |
+
"global_step": 71500,
|
| 858 |
+
"eval_loss": 16.205015182495117,
|
| 859 |
+
"eval_cosine_accuracy": 0.7656894326210022
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 1.764014112112897,
|
| 863 |
+
"global_step": 72000,
|
| 864 |
+
"eval_loss": 16.384424209594727,
|
| 865 |
+
"eval_cosine_accuracy": 0.7568535208702087
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 1.776264210113681,
|
| 869 |
+
"global_step": 72500,
|
| 870 |
+
"eval_loss": 16.72735595703125,
|
| 871 |
+
"eval_cosine_accuracy": 0.7388191819190979
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.7885143081144648,
|
| 875 |
+
"global_step": 73000,
|
| 876 |
+
"eval_loss": 15.925857543945312,
|
| 877 |
+
"eval_cosine_accuracy": 0.7689972519874573
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 1.800764406115249,
|
| 881 |
+
"global_step": 73500,
|
| 882 |
+
"eval_loss": 16.406005859375,
|
| 883 |
+
"eval_cosine_accuracy": 0.7714440822601318
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 1.8130145041160328,
|
| 887 |
+
"global_step": 74000,
|
| 888 |
+
"eval_loss": 16.443193435668945,
|
| 889 |
+
"eval_cosine_accuracy": 0.7695409655570984
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 1.825264602116817,
|
| 893 |
+
"global_step": 74500,
|
| 894 |
+
"eval_loss": 16.270015716552734,
|
| 895 |
+
"eval_cosine_accuracy": 0.7670941352844238
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 1.8375147001176009,
|
| 899 |
+
"global_step": 75000,
|
| 900 |
+
"eval_loss": 15.971381187438965,
|
| 901 |
+
"eval_cosine_accuracy": 0.7816393971443176
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 1.849764798118385,
|
| 905 |
+
"global_step": 75500,
|
| 906 |
+
"eval_loss": 16.55946922302246,
|
| 907 |
+
"eval_cosine_accuracy": 0.7499207258224487
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 1.862014896119169,
|
| 911 |
+
"global_step": 76000,
|
| 912 |
+
"eval_loss": 16.734323501586914,
|
| 913 |
+
"eval_cosine_accuracy": 0.752140998840332
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.874264994119953,
|
| 917 |
+
"global_step": 76500,
|
| 918 |
+
"eval_loss": 16.345609664916992,
|
| 919 |
+
"eval_cosine_accuracy": 0.7608410120010376
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 1.886515092120737,
|
| 923 |
+
"global_step": 77000,
|
| 924 |
+
"eval_loss": 16.739871978759766,
|
| 925 |
+
"eval_cosine_accuracy": 0.7591191530227661
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 1.898765190121521,
|
| 929 |
+
"global_step": 77500,
|
| 930 |
+
"eval_loss": 16.120464324951172,
|
| 931 |
+
"eval_cosine_accuracy": 0.7626988291740417
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 1.911015288122305,
|
| 935 |
+
"global_step": 78000,
|
| 936 |
+
"eval_loss": 16.300392150878906,
|
| 937 |
+
"eval_cosine_accuracy": 0.7634690999984741
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 1.9232653861230888,
|
| 941 |
+
"global_step": 78500,
|
| 942 |
+
"eval_loss": 16.190589904785156,
|
| 943 |
+
"eval_cosine_accuracy": 0.7540441155433655
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 1.935515484123873,
|
| 947 |
+
"global_step": 79000,
|
| 948 |
+
"eval_loss": 16.35584831237793,
|
| 949 |
+
"eval_cosine_accuracy": 0.7497847676277161
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 1.947765582124657,
|
| 953 |
+
"global_step": 79500,
|
| 954 |
+
"eval_loss": 16.15901756286621,
|
| 955 |
+
"eval_cosine_accuracy": 0.7463863492012024
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.9600156801254411,
|
| 959 |
+
"global_step": 80000,
|
| 960 |
+
"eval_loss": 16.208189010620117,
|
| 961 |
+
"eval_cosine_accuracy": 0.7588472366333008
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 1.972265778126225,
|
| 965 |
+
"global_step": 80500,
|
| 966 |
+
"eval_loss": 16.452190399169922,
|
| 967 |
+
"eval_cosine_accuracy": 0.7568082213401794
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 1.984515876127009,
|
| 971 |
+
"global_step": 81000,
|
| 972 |
+
"eval_loss": 16.443246841430664,
|
| 973 |
+
"eval_cosine_accuracy": 0.7486972808837891
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 1.996765974127793,
|
| 977 |
+
"global_step": 81500,
|
| 978 |
+
"eval_loss": 16.488916397094727,
|
| 979 |
+
"eval_cosine_accuracy": 0.7561737895011902
|
| 980 |
+
}
|
| 981 |
+
],
|
| 982 |
+
"current_epoch": 2,
|
| 983 |
+
"global_step": 81632
|
| 984 |
+
}
|
checkpoint-81632/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|