Training checkpoint - Epoch 3, Step 85563
Browse files- checkpoint-85563/1_Pooling/config.json +10 -0
- checkpoint-85563/config.json +25 -0
- checkpoint-85563/config_sentence_transformers.json +14 -0
- checkpoint-85563/model.safetensors +3 -0
- checkpoint-85563/modules.json +20 -0
- checkpoint-85563/optimizer.pt +3 -0
- checkpoint-85563/rng_state.pth +3 -0
- checkpoint-85563/scaler.pt +3 -0
- checkpoint-85563/scheduler.pt +3 -0
- checkpoint-85563/sentence_bert_config.json +4 -0
- checkpoint-85563/special_tokens_map.json +37 -0
- checkpoint-85563/tokenizer.json +0 -0
- checkpoint-85563/tokenizer_config.json +65 -0
- checkpoint-85563/trainer_state.json +2777 -0
- checkpoint-85563/training_args.bin +3 -0
- checkpoint-85563/training_metrics.json +1032 -0
- checkpoint-85563/vocab.txt +0 -0
checkpoint-85563/1_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-85563/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-85563/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-85563/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:834b1a7a008ccee80e211114a9dfb8c8958e5c0f5c20dd78a8d2ea403b425a77
|
| 3 |
+
size 90864192
|
checkpoint-85563/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-85563/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e7834a822c7801a06d0744b41595c69f40f84a9f4bc630e34a8c001a7e091a77
|
| 3 |
+
size 180607738
|
checkpoint-85563/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5f2d48debb5ce87f85e81b3b803221eb87314b3a93bc2b4c3c6110a813df490d
|
| 3 |
+
size 14244
|
checkpoint-85563/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ef0df67dceab8d8e064024b275d1a9dd6528e66d0fa8ef2dc062e76db866545b
|
| 3 |
+
size 988
|
checkpoint-85563/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:95440bffbb833898a305f56ef81cf33cf04ca5d573233972a249e9827a7378a5
|
| 3 |
+
size 1064
|
checkpoint-85563/sentence_bert_config.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"max_seq_length": 256,
|
| 3 |
+
"do_lower_case": false
|
| 4 |
+
}
|
checkpoint-85563/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-85563/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-85563/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-85563/trainer_state.json
ADDED
|
@@ -0,0 +1,2777 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 3.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 85563,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 3.5061884225658285e-05,
|
| 14 |
+
"grad_norm": NaN,
|
| 15 |
+
"learning_rate": 0.0,
|
| 16 |
+
"loss": 17.369,
|
| 17 |
+
"step": 1
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.017530942112829143,
|
| 21 |
+
"grad_norm": 69.03182983398438,
|
| 22 |
+
"learning_rate": 3.4782608695652175e-07,
|
| 23 |
+
"loss": 16.4014,
|
| 24 |
+
"step": 500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.017530942112829143,
|
| 28 |
+
"eval_cosine_accuracy": 0.9315329194068909,
|
| 29 |
+
"eval_loss": 11.649206161499023,
|
| 30 |
+
"eval_runtime": 50.3154,
|
| 31 |
+
"eval_samples_per_second": 375.948,
|
| 32 |
+
"eval_steps_per_second": 1.471,
|
| 33 |
+
"step": 500
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.035061884225658285,
|
| 37 |
+
"grad_norm": 70.3837890625,
|
| 38 |
+
"learning_rate": 6.984572230014027e-07,
|
| 39 |
+
"loss": 15.2978,
|
| 40 |
+
"step": 1000
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.035061884225658285,
|
| 44 |
+
"eval_cosine_accuracy": 0.9300376176834106,
|
| 45 |
+
"eval_loss": 11.218177795410156,
|
| 46 |
+
"eval_runtime": 51.2467,
|
| 47 |
+
"eval_samples_per_second": 369.116,
|
| 48 |
+
"eval_steps_per_second": 1.444,
|
| 49 |
+
"step": 1000
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.05259282633848743,
|
| 53 |
+
"grad_norm": 28.8201847076416,
|
| 54 |
+
"learning_rate": 1.0490883590462834e-06,
|
| 55 |
+
"loss": 12.9309,
|
| 56 |
+
"step": 1500
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.05259282633848743,
|
| 60 |
+
"eval_cosine_accuracy": 0.9178032279014587,
|
| 61 |
+
"eval_loss": 9.87814998626709,
|
| 62 |
+
"eval_runtime": 50.3509,
|
| 63 |
+
"eval_samples_per_second": 375.683,
|
| 64 |
+
"eval_steps_per_second": 1.47,
|
| 65 |
+
"step": 1500
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.07012376845131657,
|
| 69 |
+
"grad_norm": 5.389261245727539,
|
| 70 |
+
"learning_rate": 1.399719495091164e-06,
|
| 71 |
+
"loss": 10.6731,
|
| 72 |
+
"step": 2000
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.07012376845131657,
|
| 76 |
+
"eval_cosine_accuracy": 0.9117313623428345,
|
| 77 |
+
"eval_loss": 9.138021469116211,
|
| 78 |
+
"eval_runtime": 51.016,
|
| 79 |
+
"eval_samples_per_second": 370.785,
|
| 80 |
+
"eval_steps_per_second": 1.451,
|
| 81 |
+
"step": 2000
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.08765471056414571,
|
| 85 |
+
"grad_norm": 3.7415714263916016,
|
| 86 |
+
"learning_rate": 1.7503506311360449e-06,
|
| 87 |
+
"loss": 10.1212,
|
| 88 |
+
"step": 2500
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.08765471056414571,
|
| 92 |
+
"eval_cosine_accuracy": 0.9272282123565674,
|
| 93 |
+
"eval_loss": 9.000126838684082,
|
| 94 |
+
"eval_runtime": 50.3919,
|
| 95 |
+
"eval_samples_per_second": 375.378,
|
| 96 |
+
"eval_steps_per_second": 1.468,
|
| 97 |
+
"step": 2500
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.10518565267697486,
|
| 101 |
+
"grad_norm": 3.660752058029175,
|
| 102 |
+
"learning_rate": 2.1009817671809257e-06,
|
| 103 |
+
"loss": 10.0443,
|
| 104 |
+
"step": 3000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.10518565267697486,
|
| 108 |
+
"eval_cosine_accuracy": 0.9307625889778137,
|
| 109 |
+
"eval_loss": 8.966527938842773,
|
| 110 |
+
"eval_runtime": 50.9518,
|
| 111 |
+
"eval_samples_per_second": 371.253,
|
| 112 |
+
"eval_steps_per_second": 1.452,
|
| 113 |
+
"step": 3000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.122716594789804,
|
| 117 |
+
"grad_norm": 3.7356393337249756,
|
| 118 |
+
"learning_rate": 2.4516129032258066e-06,
|
| 119 |
+
"loss": 9.9987,
|
| 120 |
+
"step": 3500
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.122716594789804,
|
| 124 |
+
"eval_cosine_accuracy": 0.9352032542228699,
|
| 125 |
+
"eval_loss": 8.893427848815918,
|
| 126 |
+
"eval_runtime": 50.4527,
|
| 127 |
+
"eval_samples_per_second": 374.925,
|
| 128 |
+
"eval_steps_per_second": 1.467,
|
| 129 |
+
"step": 3500
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.14024753690263314,
|
| 133 |
+
"grad_norm": 3.114379644393921,
|
| 134 |
+
"learning_rate": 2.8022440392706874e-06,
|
| 135 |
+
"loss": 9.9643,
|
| 136 |
+
"step": 4000
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.14024753690263314,
|
| 140 |
+
"eval_cosine_accuracy": 0.9365172982215881,
|
| 141 |
+
"eval_loss": 8.888650894165039,
|
| 142 |
+
"eval_runtime": 51.3556,
|
| 143 |
+
"eval_samples_per_second": 368.334,
|
| 144 |
+
"eval_steps_per_second": 1.441,
|
| 145 |
+
"step": 4000
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.1577784790154623,
|
| 149 |
+
"grad_norm": 3.3485982418060303,
|
| 150 |
+
"learning_rate": 3.1528751753155682e-06,
|
| 151 |
+
"loss": 9.9351,
|
| 152 |
+
"step": 4500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.1577784790154623,
|
| 156 |
+
"eval_cosine_accuracy": 0.9367438554763794,
|
| 157 |
+
"eval_loss": 8.871306419372559,
|
| 158 |
+
"eval_runtime": 50.3996,
|
| 159 |
+
"eval_samples_per_second": 375.32,
|
| 160 |
+
"eval_steps_per_second": 1.468,
|
| 161 |
+
"step": 4500
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.17530942112829143,
|
| 165 |
+
"grad_norm": 3.7280542850494385,
|
| 166 |
+
"learning_rate": 3.503506311360449e-06,
|
| 167 |
+
"loss": 9.9138,
|
| 168 |
+
"step": 5000
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 0.17530942112829143,
|
| 172 |
+
"eval_cosine_accuracy": 0.940504789352417,
|
| 173 |
+
"eval_loss": 8.830900192260742,
|
| 174 |
+
"eval_runtime": 51.0146,
|
| 175 |
+
"eval_samples_per_second": 370.796,
|
| 176 |
+
"eval_steps_per_second": 1.451,
|
| 177 |
+
"step": 5000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.19284036324112058,
|
| 181 |
+
"grad_norm": 3.35213303565979,
|
| 182 |
+
"learning_rate": 3.85413744740533e-06,
|
| 183 |
+
"loss": 9.8921,
|
| 184 |
+
"step": 5500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.19284036324112058,
|
| 188 |
+
"eval_cosine_accuracy": 0.9402329325675964,
|
| 189 |
+
"eval_loss": 8.825165748596191,
|
| 190 |
+
"eval_runtime": 50.5432,
|
| 191 |
+
"eval_samples_per_second": 374.254,
|
| 192 |
+
"eval_steps_per_second": 1.464,
|
| 193 |
+
"step": 5500
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.2103713053539497,
|
| 197 |
+
"grad_norm": 3.040640115737915,
|
| 198 |
+
"learning_rate": 4.204768583450211e-06,
|
| 199 |
+
"loss": 9.8705,
|
| 200 |
+
"step": 6000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.2103713053539497,
|
| 204 |
+
"eval_cosine_accuracy": 0.941592276096344,
|
| 205 |
+
"eval_loss": 8.791216850280762,
|
| 206 |
+
"eval_runtime": 50.1354,
|
| 207 |
+
"eval_samples_per_second": 377.299,
|
| 208 |
+
"eval_steps_per_second": 1.476,
|
| 209 |
+
"step": 6000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.22790224746677887,
|
| 213 |
+
"grad_norm": 4.0565409660339355,
|
| 214 |
+
"learning_rate": 4.555399719495092e-06,
|
| 215 |
+
"loss": 9.8535,
|
| 216 |
+
"step": 6500
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.22790224746677887,
|
| 220 |
+
"eval_cosine_accuracy": 0.9428157210350037,
|
| 221 |
+
"eval_loss": 8.7944917678833,
|
| 222 |
+
"eval_runtime": 51.4587,
|
| 223 |
+
"eval_samples_per_second": 367.596,
|
| 224 |
+
"eval_steps_per_second": 1.438,
|
| 225 |
+
"step": 6500
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.245433189579608,
|
| 229 |
+
"grad_norm": 3.325587272644043,
|
| 230 |
+
"learning_rate": 4.906030855539972e-06,
|
| 231 |
+
"loss": 9.8412,
|
| 232 |
+
"step": 7000
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.245433189579608,
|
| 236 |
+
"eval_cosine_accuracy": 0.9449000954627991,
|
| 237 |
+
"eval_loss": 8.775042533874512,
|
| 238 |
+
"eval_runtime": 51.5544,
|
| 239 |
+
"eval_samples_per_second": 366.914,
|
| 240 |
+
"eval_steps_per_second": 1.435,
|
| 241 |
+
"step": 7000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.26296413169243715,
|
| 245 |
+
"grad_norm": 3.436500072479248,
|
| 246 |
+
"learning_rate": 5.256661991584852e-06,
|
| 247 |
+
"loss": 9.8243,
|
| 248 |
+
"step": 7500
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.26296413169243715,
|
| 252 |
+
"eval_cosine_accuracy": 0.9455344676971436,
|
| 253 |
+
"eval_loss": 8.7813720703125,
|
| 254 |
+
"eval_runtime": 50.6094,
|
| 255 |
+
"eval_samples_per_second": 373.764,
|
| 256 |
+
"eval_steps_per_second": 1.462,
|
| 257 |
+
"step": 7500
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.2804950738052663,
|
| 261 |
+
"grad_norm": 4.005023002624512,
|
| 262 |
+
"learning_rate": 5.607293127629734e-06,
|
| 263 |
+
"loss": 9.8144,
|
| 264 |
+
"step": 8000
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.2804950738052663,
|
| 268 |
+
"eval_cosine_accuracy": 0.9440391659736633,
|
| 269 |
+
"eval_loss": 8.798937797546387,
|
| 270 |
+
"eval_runtime": 50.431,
|
| 271 |
+
"eval_samples_per_second": 375.087,
|
| 272 |
+
"eval_steps_per_second": 1.467,
|
| 273 |
+
"step": 8000
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.29802601591809547,
|
| 277 |
+
"grad_norm": 3.6476752758026123,
|
| 278 |
+
"learning_rate": 5.957924263674615e-06,
|
| 279 |
+
"loss": 9.7972,
|
| 280 |
+
"step": 8500
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.29802601591809547,
|
| 284 |
+
"eval_cosine_accuracy": 0.947256326675415,
|
| 285 |
+
"eval_loss": 8.76003360748291,
|
| 286 |
+
"eval_runtime": 51.8469,
|
| 287 |
+
"eval_samples_per_second": 364.843,
|
| 288 |
+
"eval_steps_per_second": 1.427,
|
| 289 |
+
"step": 8500
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.3155569580309246,
|
| 293 |
+
"grad_norm": 3.4247946739196777,
|
| 294 |
+
"learning_rate": 6.308555399719496e-06,
|
| 295 |
+
"loss": 9.7881,
|
| 296 |
+
"step": 9000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.3155569580309246,
|
| 300 |
+
"eval_cosine_accuracy": 0.9472110271453857,
|
| 301 |
+
"eval_loss": 8.755824089050293,
|
| 302 |
+
"eval_runtime": 51.2026,
|
| 303 |
+
"eval_samples_per_second": 369.435,
|
| 304 |
+
"eval_steps_per_second": 1.445,
|
| 305 |
+
"step": 9000
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.3330879001437537,
|
| 309 |
+
"grad_norm": 3.730705738067627,
|
| 310 |
+
"learning_rate": 6.659186535764376e-06,
|
| 311 |
+
"loss": 9.7692,
|
| 312 |
+
"step": 9500
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 0.3330879001437537,
|
| 316 |
+
"eval_cosine_accuracy": 0.9468938112258911,
|
| 317 |
+
"eval_loss": 8.768818855285645,
|
| 318 |
+
"eval_runtime": 50.5549,
|
| 319 |
+
"eval_samples_per_second": 374.168,
|
| 320 |
+
"eval_steps_per_second": 1.464,
|
| 321 |
+
"step": 9500
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.35061884225658285,
|
| 325 |
+
"grad_norm": 4.8609819412231445,
|
| 326 |
+
"learning_rate": 7.009817671809257e-06,
|
| 327 |
+
"loss": 9.7628,
|
| 328 |
+
"step": 10000
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.35061884225658285,
|
| 332 |
+
"eval_cosine_accuracy": 0.9488422870635986,
|
| 333 |
+
"eval_loss": 8.746337890625,
|
| 334 |
+
"eval_runtime": 51.129,
|
| 335 |
+
"eval_samples_per_second": 369.966,
|
| 336 |
+
"eval_steps_per_second": 1.447,
|
| 337 |
+
"step": 10000
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 0.36814978436941204,
|
| 341 |
+
"grad_norm": 3.147473096847534,
|
| 342 |
+
"learning_rate": 7.359747545582048e-06,
|
| 343 |
+
"loss": 9.755,
|
| 344 |
+
"step": 10500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.36814978436941204,
|
| 348 |
+
"eval_cosine_accuracy": 0.9490688443183899,
|
| 349 |
+
"eval_loss": 8.786169052124023,
|
| 350 |
+
"eval_runtime": 50.5305,
|
| 351 |
+
"eval_samples_per_second": 374.348,
|
| 352 |
+
"eval_steps_per_second": 1.464,
|
| 353 |
+
"step": 10500
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.38568072648224117,
|
| 357 |
+
"grad_norm": 3.352139711380005,
|
| 358 |
+
"learning_rate": 7.710378681626928e-06,
|
| 359 |
+
"loss": 9.7403,
|
| 360 |
+
"step": 11000
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.38568072648224117,
|
| 364 |
+
"eval_cosine_accuracy": 0.9479813575744629,
|
| 365 |
+
"eval_loss": 8.785058975219727,
|
| 366 |
+
"eval_runtime": 51.154,
|
| 367 |
+
"eval_samples_per_second": 369.785,
|
| 368 |
+
"eval_steps_per_second": 1.447,
|
| 369 |
+
"step": 11000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.4032116685950703,
|
| 373 |
+
"grad_norm": 3.792637586593628,
|
| 374 |
+
"learning_rate": 8.06100981767181e-06,
|
| 375 |
+
"loss": 9.7309,
|
| 376 |
+
"step": 11500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 0.4032116685950703,
|
| 380 |
+
"eval_cosine_accuracy": 0.9496579170227051,
|
| 381 |
+
"eval_loss": 8.779532432556152,
|
| 382 |
+
"eval_runtime": 50.5958,
|
| 383 |
+
"eval_samples_per_second": 373.865,
|
| 384 |
+
"eval_steps_per_second": 1.463,
|
| 385 |
+
"step": 11500
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 0.4207426107078994,
|
| 389 |
+
"grad_norm": 3.8172216415405273,
|
| 390 |
+
"learning_rate": 8.411640953716691e-06,
|
| 391 |
+
"loss": 9.7211,
|
| 392 |
+
"step": 12000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.4207426107078994,
|
| 396 |
+
"eval_cosine_accuracy": 0.9478000998497009,
|
| 397 |
+
"eval_loss": 8.792497634887695,
|
| 398 |
+
"eval_runtime": 51.0924,
|
| 399 |
+
"eval_samples_per_second": 370.231,
|
| 400 |
+
"eval_steps_per_second": 1.448,
|
| 401 |
+
"step": 12000
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.4382735528207286,
|
| 405 |
+
"grad_norm": 4.221671104431152,
|
| 406 |
+
"learning_rate": 8.761570827489482e-06,
|
| 407 |
+
"loss": 9.71,
|
| 408 |
+
"step": 12500
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.4382735528207286,
|
| 412 |
+
"eval_cosine_accuracy": 0.949793815612793,
|
| 413 |
+
"eval_loss": 8.789593696594238,
|
| 414 |
+
"eval_runtime": 50.2993,
|
| 415 |
+
"eval_samples_per_second": 376.069,
|
| 416 |
+
"eval_steps_per_second": 1.471,
|
| 417 |
+
"step": 12500
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.45580449493355774,
|
| 421 |
+
"grad_norm": 3.5044078826904297,
|
| 422 |
+
"learning_rate": 9.112201963534364e-06,
|
| 423 |
+
"loss": 9.7043,
|
| 424 |
+
"step": 13000
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 0.45580449493355774,
|
| 428 |
+
"eval_cosine_accuracy": 0.9480266571044922,
|
| 429 |
+
"eval_loss": 8.827651977539062,
|
| 430 |
+
"eval_runtime": 50.4208,
|
| 431 |
+
"eval_samples_per_second": 375.162,
|
| 432 |
+
"eval_steps_per_second": 1.468,
|
| 433 |
+
"step": 13000
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.47333543704638686,
|
| 437 |
+
"grad_norm": 3.464083194732666,
|
| 438 |
+
"learning_rate": 9.462833099579244e-06,
|
| 439 |
+
"loss": 9.6925,
|
| 440 |
+
"step": 13500
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.47333543704638686,
|
| 444 |
+
"eval_cosine_accuracy": 0.9472110271453857,
|
| 445 |
+
"eval_loss": 8.85822868347168,
|
| 446 |
+
"eval_runtime": 52.1141,
|
| 447 |
+
"eval_samples_per_second": 362.972,
|
| 448 |
+
"eval_steps_per_second": 1.42,
|
| 449 |
+
"step": 13500
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.490866379159216,
|
| 453 |
+
"grad_norm": 3.7905664443969727,
|
| 454 |
+
"learning_rate": 9.813464235624124e-06,
|
| 455 |
+
"loss": 9.6877,
|
| 456 |
+
"step": 14000
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 0.490866379159216,
|
| 460 |
+
"eval_cosine_accuracy": 0.9473016262054443,
|
| 461 |
+
"eval_loss": 8.843923568725586,
|
| 462 |
+
"eval_runtime": 51.4524,
|
| 463 |
+
"eval_samples_per_second": 367.641,
|
| 464 |
+
"eval_steps_per_second": 1.438,
|
| 465 |
+
"step": 14000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.5083973212720452,
|
| 469 |
+
"grad_norm": 4.104105472564697,
|
| 470 |
+
"learning_rate": 1.0163394109396916e-05,
|
| 471 |
+
"loss": 9.6751,
|
| 472 |
+
"step": 14500
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.5083973212720452,
|
| 476 |
+
"eval_cosine_accuracy": 0.9467579126358032,
|
| 477 |
+
"eval_loss": 8.850715637207031,
|
| 478 |
+
"eval_runtime": 50.7244,
|
| 479 |
+
"eval_samples_per_second": 372.917,
|
| 480 |
+
"eval_steps_per_second": 1.459,
|
| 481 |
+
"step": 14500
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 0.5259282633848743,
|
| 485 |
+
"grad_norm": 3.7694902420043945,
|
| 486 |
+
"learning_rate": 1.0514025245441798e-05,
|
| 487 |
+
"loss": 9.6727,
|
| 488 |
+
"step": 15000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.5259282633848743,
|
| 492 |
+
"eval_cosine_accuracy": 0.9481172561645508,
|
| 493 |
+
"eval_loss": 8.862038612365723,
|
| 494 |
+
"eval_runtime": 51.5121,
|
| 495 |
+
"eval_samples_per_second": 367.215,
|
| 496 |
+
"eval_steps_per_second": 1.437,
|
| 497 |
+
"step": 15000
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 0.5434592054977034,
|
| 501 |
+
"grad_norm": 4.85964298248291,
|
| 502 |
+
"learning_rate": 1.0864656381486678e-05,
|
| 503 |
+
"loss": 9.6595,
|
| 504 |
+
"step": 15500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.5434592054977034,
|
| 508 |
+
"eval_cosine_accuracy": 0.9488875865936279,
|
| 509 |
+
"eval_loss": 8.83780574798584,
|
| 510 |
+
"eval_runtime": 50.6086,
|
| 511 |
+
"eval_samples_per_second": 373.771,
|
| 512 |
+
"eval_steps_per_second": 1.462,
|
| 513 |
+
"step": 15500
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.5609901476105326,
|
| 517 |
+
"grad_norm": 3.9702301025390625,
|
| 518 |
+
"learning_rate": 1.1215287517531558e-05,
|
| 519 |
+
"loss": 9.6573,
|
| 520 |
+
"step": 16000
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.5609901476105326,
|
| 524 |
+
"eval_cosine_accuracy": 0.9463953971862793,
|
| 525 |
+
"eval_loss": 8.849825859069824,
|
| 526 |
+
"eval_runtime": 51.1328,
|
| 527 |
+
"eval_samples_per_second": 369.939,
|
| 528 |
+
"eval_steps_per_second": 1.447,
|
| 529 |
+
"step": 16000
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.5785210897233617,
|
| 533 |
+
"grad_norm": 4.0049519538879395,
|
| 534 |
+
"learning_rate": 1.1565217391304348e-05,
|
| 535 |
+
"loss": 9.6451,
|
| 536 |
+
"step": 16500
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.5785210897233617,
|
| 540 |
+
"eval_cosine_accuracy": 0.9481625556945801,
|
| 541 |
+
"eval_loss": 8.855111122131348,
|
| 542 |
+
"eval_runtime": 50.7845,
|
| 543 |
+
"eval_samples_per_second": 372.476,
|
| 544 |
+
"eval_steps_per_second": 1.457,
|
| 545 |
+
"step": 16500
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 0.5960520318361909,
|
| 549 |
+
"grad_norm": 3.9873130321502686,
|
| 550 |
+
"learning_rate": 1.191584852734923e-05,
|
| 551 |
+
"loss": 9.6413,
|
| 552 |
+
"step": 17000
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 0.5960520318361909,
|
| 556 |
+
"eval_cosine_accuracy": 0.9444469809532166,
|
| 557 |
+
"eval_loss": 8.909035682678223,
|
| 558 |
+
"eval_runtime": 50.3477,
|
| 559 |
+
"eval_samples_per_second": 375.708,
|
| 560 |
+
"eval_steps_per_second": 1.47,
|
| 561 |
+
"step": 17000
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 0.6135829739490201,
|
| 565 |
+
"grad_norm": 5.137877941131592,
|
| 566 |
+
"learning_rate": 1.2265778401122022e-05,
|
| 567 |
+
"loss": 9.6343,
|
| 568 |
+
"step": 17500
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 0.6135829739490201,
|
| 572 |
+
"eval_cosine_accuracy": 0.9443563222885132,
|
| 573 |
+
"eval_loss": 8.95804500579834,
|
| 574 |
+
"eval_runtime": 51.2881,
|
| 575 |
+
"eval_samples_per_second": 368.819,
|
| 576 |
+
"eval_steps_per_second": 1.443,
|
| 577 |
+
"step": 17500
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.6311139160618492,
|
| 581 |
+
"grad_norm": 5.1459174156188965,
|
| 582 |
+
"learning_rate": 1.26164095371669e-05,
|
| 583 |
+
"loss": 9.6284,
|
| 584 |
+
"step": 18000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.6311139160618492,
|
| 588 |
+
"eval_cosine_accuracy": 0.9461235404014587,
|
| 589 |
+
"eval_loss": 8.915170669555664,
|
| 590 |
+
"eval_runtime": 51.0383,
|
| 591 |
+
"eval_samples_per_second": 370.623,
|
| 592 |
+
"eval_steps_per_second": 1.45,
|
| 593 |
+
"step": 18000
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 0.6486448581746783,
|
| 597 |
+
"grad_norm": 5.0631794929504395,
|
| 598 |
+
"learning_rate": 1.2967040673211782e-05,
|
| 599 |
+
"loss": 9.6219,
|
| 600 |
+
"step": 18500
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 0.6486448581746783,
|
| 604 |
+
"eval_cosine_accuracy": 0.9455344676971436,
|
| 605 |
+
"eval_loss": 8.9291353225708,
|
| 606 |
+
"eval_runtime": 51.9477,
|
| 607 |
+
"eval_samples_per_second": 364.135,
|
| 608 |
+
"eval_steps_per_second": 1.425,
|
| 609 |
+
"step": 18500
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.6661758002875074,
|
| 613 |
+
"grad_norm": 3.612875461578369,
|
| 614 |
+
"learning_rate": 1.3317671809256662e-05,
|
| 615 |
+
"loss": 9.6129,
|
| 616 |
+
"step": 19000
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 0.6661758002875074,
|
| 620 |
+
"eval_cosine_accuracy": 0.9464406967163086,
|
| 621 |
+
"eval_loss": 8.916707038879395,
|
| 622 |
+
"eval_runtime": 52.0312,
|
| 623 |
+
"eval_samples_per_second": 363.551,
|
| 624 |
+
"eval_steps_per_second": 1.422,
|
| 625 |
+
"step": 19000
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.6837067424003366,
|
| 629 |
+
"grad_norm": 3.58630108833313,
|
| 630 |
+
"learning_rate": 1.3668302945301544e-05,
|
| 631 |
+
"loss": 9.6086,
|
| 632 |
+
"step": 19500
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.6837067424003366,
|
| 636 |
+
"eval_cosine_accuracy": 0.9478000998497009,
|
| 637 |
+
"eval_loss": 8.885634422302246,
|
| 638 |
+
"eval_runtime": 51.0372,
|
| 639 |
+
"eval_samples_per_second": 370.631,
|
| 640 |
+
"eval_steps_per_second": 1.45,
|
| 641 |
+
"step": 19500
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.7012376845131657,
|
| 645 |
+
"grad_norm": 4.414312839508057,
|
| 646 |
+
"learning_rate": 1.4018934081346425e-05,
|
| 647 |
+
"loss": 9.6,
|
| 648 |
+
"step": 20000
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 0.7012376845131657,
|
| 652 |
+
"eval_cosine_accuracy": 0.9428157210350037,
|
| 653 |
+
"eval_loss": 9.033989906311035,
|
| 654 |
+
"eval_runtime": 51.5213,
|
| 655 |
+
"eval_samples_per_second": 367.149,
|
| 656 |
+
"eval_steps_per_second": 1.436,
|
| 657 |
+
"step": 20000
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 0.7187686266259948,
|
| 661 |
+
"grad_norm": 4.458536148071289,
|
| 662 |
+
"learning_rate": 1.4369565217391305e-05,
|
| 663 |
+
"loss": 9.595,
|
| 664 |
+
"step": 20500
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 0.7187686266259948,
|
| 668 |
+
"eval_cosine_accuracy": 0.9464859962463379,
|
| 669 |
+
"eval_loss": 8.914213180541992,
|
| 670 |
+
"eval_runtime": 50.8951,
|
| 671 |
+
"eval_samples_per_second": 371.666,
|
| 672 |
+
"eval_steps_per_second": 1.454,
|
| 673 |
+
"step": 20500
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 0.7362995687388241,
|
| 677 |
+
"grad_norm": 3.8019843101501465,
|
| 678 |
+
"learning_rate": 1.4720196353436187e-05,
|
| 679 |
+
"loss": 9.5897,
|
| 680 |
+
"step": 21000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.7362995687388241,
|
| 684 |
+
"eval_cosine_accuracy": 0.9437219500541687,
|
| 685 |
+
"eval_loss": 9.02304458618164,
|
| 686 |
+
"eval_runtime": 51.2548,
|
| 687 |
+
"eval_samples_per_second": 369.058,
|
| 688 |
+
"eval_steps_per_second": 1.444,
|
| 689 |
+
"step": 21000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.7538305108516532,
|
| 693 |
+
"grad_norm": 4.581631183624268,
|
| 694 |
+
"learning_rate": 1.5070827489481069e-05,
|
| 695 |
+
"loss": 9.5815,
|
| 696 |
+
"step": 21500
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.7538305108516532,
|
| 700 |
+
"eval_cosine_accuracy": 0.9458063244819641,
|
| 701 |
+
"eval_loss": 8.924837112426758,
|
| 702 |
+
"eval_runtime": 52.5092,
|
| 703 |
+
"eval_samples_per_second": 360.242,
|
| 704 |
+
"eval_steps_per_second": 1.409,
|
| 705 |
+
"step": 21500
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 0.7713614529644823,
|
| 709 |
+
"grad_norm": 4.277956962585449,
|
| 710 |
+
"learning_rate": 1.5420757363253856e-05,
|
| 711 |
+
"loss": 9.5783,
|
| 712 |
+
"step": 22000
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 0.7713614529644823,
|
| 716 |
+
"eval_cosine_accuracy": 0.9445375800132751,
|
| 717 |
+
"eval_loss": 8.93367862701416,
|
| 718 |
+
"eval_runtime": 51.3915,
|
| 719 |
+
"eval_samples_per_second": 368.077,
|
| 720 |
+
"eval_steps_per_second": 1.44,
|
| 721 |
+
"step": 22000
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 0.7888923950773115,
|
| 725 |
+
"grad_norm": 4.0991034507751465,
|
| 726 |
+
"learning_rate": 1.577138849929874e-05,
|
| 727 |
+
"loss": 9.5692,
|
| 728 |
+
"step": 22500
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 0.7888923950773115,
|
| 732 |
+
"eval_cosine_accuracy": 0.9451719522476196,
|
| 733 |
+
"eval_loss": 8.977583885192871,
|
| 734 |
+
"eval_runtime": 51.0271,
|
| 735 |
+
"eval_samples_per_second": 370.705,
|
| 736 |
+
"eval_steps_per_second": 1.45,
|
| 737 |
+
"step": 22500
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.8064233371901406,
|
| 741 |
+
"grad_norm": 4.957141399383545,
|
| 742 |
+
"learning_rate": 1.612201963534362e-05,
|
| 743 |
+
"loss": 9.5637,
|
| 744 |
+
"step": 23000
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.8064233371901406,
|
| 748 |
+
"eval_cosine_accuracy": 0.9423173069953918,
|
| 749 |
+
"eval_loss": 9.053213119506836,
|
| 750 |
+
"eval_runtime": 51.6679,
|
| 751 |
+
"eval_samples_per_second": 366.108,
|
| 752 |
+
"eval_steps_per_second": 1.432,
|
| 753 |
+
"step": 23000
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.8239542793029697,
|
| 757 |
+
"grad_norm": 4.1165876388549805,
|
| 758 |
+
"learning_rate": 1.64726507713885e-05,
|
| 759 |
+
"loss": 9.5631,
|
| 760 |
+
"step": 23500
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 0.8239542793029697,
|
| 764 |
+
"eval_cosine_accuracy": 0.9439032077789307,
|
| 765 |
+
"eval_loss": 9.02741527557373,
|
| 766 |
+
"eval_runtime": 50.7204,
|
| 767 |
+
"eval_samples_per_second": 372.947,
|
| 768 |
+
"eval_steps_per_second": 1.459,
|
| 769 |
+
"step": 23500
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 0.8414852214157988,
|
| 773 |
+
"grad_norm": 3.1949024200439453,
|
| 774 |
+
"learning_rate": 1.68218793828892e-05,
|
| 775 |
+
"loss": 9.553,
|
| 776 |
+
"step": 24000
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 0.8414852214157988,
|
| 780 |
+
"eval_cosine_accuracy": 0.9431782364845276,
|
| 781 |
+
"eval_loss": 9.003716468811035,
|
| 782 |
+
"eval_runtime": 51.597,
|
| 783 |
+
"eval_samples_per_second": 366.61,
|
| 784 |
+
"eval_steps_per_second": 1.434,
|
| 785 |
+
"step": 24000
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 0.859016163528628,
|
| 789 |
+
"grad_norm": 3.9427781105041504,
|
| 790 |
+
"learning_rate": 1.7172510518934084e-05,
|
| 791 |
+
"loss": 9.5471,
|
| 792 |
+
"step": 24500
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 0.859016163528628,
|
| 796 |
+
"eval_cosine_accuracy": 0.9453079104423523,
|
| 797 |
+
"eval_loss": 8.999176025390625,
|
| 798 |
+
"eval_runtime": 50.836,
|
| 799 |
+
"eval_samples_per_second": 372.099,
|
| 800 |
+
"eval_steps_per_second": 1.456,
|
| 801 |
+
"step": 24500
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.8765471056414572,
|
| 805 |
+
"grad_norm": 3.976271867752075,
|
| 806 |
+
"learning_rate": 1.7523141654978964e-05,
|
| 807 |
+
"loss": 9.5477,
|
| 808 |
+
"step": 25000
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.8765471056414572,
|
| 812 |
+
"eval_cosine_accuracy": 0.9445375800132751,
|
| 813 |
+
"eval_loss": 9.014289855957031,
|
| 814 |
+
"eval_runtime": 52.0697,
|
| 815 |
+
"eval_samples_per_second": 363.282,
|
| 816 |
+
"eval_steps_per_second": 1.421,
|
| 817 |
+
"step": 25000
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 0.8940780477542863,
|
| 821 |
+
"grad_norm": 3.8886728286743164,
|
| 822 |
+
"learning_rate": 1.7873772791023844e-05,
|
| 823 |
+
"loss": 9.5425,
|
| 824 |
+
"step": 25500
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 0.8940780477542863,
|
| 828 |
+
"eval_cosine_accuracy": 0.9419094920158386,
|
| 829 |
+
"eval_loss": 9.024247169494629,
|
| 830 |
+
"eval_runtime": 50.839,
|
| 831 |
+
"eval_samples_per_second": 372.077,
|
| 832 |
+
"eval_steps_per_second": 1.456,
|
| 833 |
+
"step": 25500
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 0.9116089898671155,
|
| 837 |
+
"grad_norm": 4.732585906982422,
|
| 838 |
+
"learning_rate": 1.8224403927068727e-05,
|
| 839 |
+
"loss": 9.5359,
|
| 840 |
+
"step": 26000
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 0.9116089898671155,
|
| 844 |
+
"eval_cosine_accuracy": 0.9403688311576843,
|
| 845 |
+
"eval_loss": 9.146790504455566,
|
| 846 |
+
"eval_runtime": 51.023,
|
| 847 |
+
"eval_samples_per_second": 370.735,
|
| 848 |
+
"eval_steps_per_second": 1.45,
|
| 849 |
+
"step": 26000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.9291399319799446,
|
| 853 |
+
"grad_norm": 4.0088300704956055,
|
| 854 |
+
"learning_rate": 1.8575035063113607e-05,
|
| 855 |
+
"loss": 9.5274,
|
| 856 |
+
"step": 26500
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.9291399319799446,
|
| 860 |
+
"eval_cosine_accuracy": 0.9405500888824463,
|
| 861 |
+
"eval_loss": 9.119534492492676,
|
| 862 |
+
"eval_runtime": 49.8899,
|
| 863 |
+
"eval_samples_per_second": 379.155,
|
| 864 |
+
"eval_steps_per_second": 1.483,
|
| 865 |
+
"step": 26500
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 0.9466708740927737,
|
| 869 |
+
"grad_norm": 4.282976150512695,
|
| 870 |
+
"learning_rate": 1.8925666199158487e-05,
|
| 871 |
+
"loss": 9.5226,
|
| 872 |
+
"step": 27000
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 0.9466708740927737,
|
| 876 |
+
"eval_cosine_accuracy": 0.9424079060554504,
|
| 877 |
+
"eval_loss": 9.084246635437012,
|
| 878 |
+
"eval_runtime": 49.7346,
|
| 879 |
+
"eval_samples_per_second": 380.339,
|
| 880 |
+
"eval_steps_per_second": 1.488,
|
| 881 |
+
"step": 27000
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 0.9642018162056029,
|
| 885 |
+
"grad_norm": 3.6183669567108154,
|
| 886 |
+
"learning_rate": 1.9276297335203367e-05,
|
| 887 |
+
"loss": 9.5198,
|
| 888 |
+
"step": 27500
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 0.9642018162056029,
|
| 892 |
+
"eval_cosine_accuracy": 0.9432688355445862,
|
| 893 |
+
"eval_loss": 9.046530723571777,
|
| 894 |
+
"eval_runtime": 50.9218,
|
| 895 |
+
"eval_samples_per_second": 371.471,
|
| 896 |
+
"eval_steps_per_second": 1.453,
|
| 897 |
+
"step": 27500
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 0.981732758318432,
|
| 901 |
+
"grad_norm": 4.634382247924805,
|
| 902 |
+
"learning_rate": 1.962622720897616e-05,
|
| 903 |
+
"loss": 9.5156,
|
| 904 |
+
"step": 28000
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 0.981732758318432,
|
| 908 |
+
"eval_cosine_accuracy": 0.9421360492706299,
|
| 909 |
+
"eval_loss": 9.10672664642334,
|
| 910 |
+
"eval_runtime": 50.621,
|
| 911 |
+
"eval_samples_per_second": 373.679,
|
| 912 |
+
"eval_steps_per_second": 1.462,
|
| 913 |
+
"step": 28000
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.9992637004312612,
|
| 917 |
+
"grad_norm": 4.65955924987793,
|
| 918 |
+
"learning_rate": 1.997685834502104e-05,
|
| 919 |
+
"loss": 9.511,
|
| 920 |
+
"step": 28500
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.9992637004312612,
|
| 924 |
+
"eval_cosine_accuracy": 0.9398251175880432,
|
| 925 |
+
"eval_loss": 9.120959281921387,
|
| 926 |
+
"eval_runtime": 49.9174,
|
| 927 |
+
"eval_samples_per_second": 378.946,
|
| 928 |
+
"eval_steps_per_second": 1.482,
|
| 929 |
+
"step": 28500
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 1.0167946425440904,
|
| 933 |
+
"grad_norm": 3.7407803535461426,
|
| 934 |
+
"learning_rate": 1.999853002241508e-05,
|
| 935 |
+
"loss": 9.4953,
|
| 936 |
+
"step": 29000
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 1.0167946425440904,
|
| 940 |
+
"eval_cosine_accuracy": 0.941592276096344,
|
| 941 |
+
"eval_loss": 9.088321685791016,
|
| 942 |
+
"eval_runtime": 50.0375,
|
| 943 |
+
"eval_samples_per_second": 378.036,
|
| 944 |
+
"eval_steps_per_second": 1.479,
|
| 945 |
+
"step": 29000
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 1.0343255846569195,
|
| 949 |
+
"grad_norm": 5.06452751159668,
|
| 950 |
+
"learning_rate": 1.99936977602965e-05,
|
| 951 |
+
"loss": 9.4928,
|
| 952 |
+
"step": 29500
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 1.0343255846569195,
|
| 956 |
+
"eval_cosine_accuracy": 0.9386922717094421,
|
| 957 |
+
"eval_loss": 9.25493049621582,
|
| 958 |
+
"eval_runtime": 51.1278,
|
| 959 |
+
"eval_samples_per_second": 369.975,
|
| 960 |
+
"eval_steps_per_second": 1.447,
|
| 961 |
+
"step": 29500
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 1.0518565267697486,
|
| 965 |
+
"grad_norm": 5.7913031578063965,
|
| 966 |
+
"learning_rate": 1.9985517261976804e-05,
|
| 967 |
+
"loss": 9.4912,
|
| 968 |
+
"step": 30000
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 1.0518565267697486,
|
| 972 |
+
"eval_cosine_accuracy": 0.9394626021385193,
|
| 973 |
+
"eval_loss": 9.212674140930176,
|
| 974 |
+
"eval_runtime": 49.8963,
|
| 975 |
+
"eval_samples_per_second": 379.107,
|
| 976 |
+
"eval_steps_per_second": 1.483,
|
| 977 |
+
"step": 30000
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 1.0693874688825777,
|
| 981 |
+
"grad_norm": 3.8804171085357666,
|
| 982 |
+
"learning_rate": 1.9973958497736584e-05,
|
| 983 |
+
"loss": 9.4849,
|
| 984 |
+
"step": 30500
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 1.0693874688825777,
|
| 988 |
+
"eval_cosine_accuracy": 0.9390094876289368,
|
| 989 |
+
"eval_loss": 9.155335426330566,
|
| 990 |
+
"eval_runtime": 51.2309,
|
| 991 |
+
"eval_samples_per_second": 369.231,
|
| 992 |
+
"eval_steps_per_second": 1.444,
|
| 993 |
+
"step": 30500
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 1.0869184109954069,
|
| 997 |
+
"grad_norm": 4.496101379394531,
|
| 998 |
+
"learning_rate": 1.995903838953484e-05,
|
| 999 |
+
"loss": 9.4787,
|
| 1000 |
+
"step": 31000
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 1.0869184109954069,
|
| 1004 |
+
"eval_cosine_accuracy": 0.9399157166481018,
|
| 1005 |
+
"eval_loss": 9.130995750427246,
|
| 1006 |
+
"eval_runtime": 49.9212,
|
| 1007 |
+
"eval_samples_per_second": 378.917,
|
| 1008 |
+
"eval_steps_per_second": 1.482,
|
| 1009 |
+
"step": 31000
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 1.104449353108236,
|
| 1013 |
+
"grad_norm": 4.565690040588379,
|
| 1014 |
+
"learning_rate": 1.9940761965627467e-05,
|
| 1015 |
+
"loss": 9.478,
|
| 1016 |
+
"step": 31500
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 1.104449353108236,
|
| 1020 |
+
"eval_cosine_accuracy": 0.9375141859054565,
|
| 1021 |
+
"eval_loss": 9.160995483398438,
|
| 1022 |
+
"eval_runtime": 50.8912,
|
| 1023 |
+
"eval_samples_per_second": 371.695,
|
| 1024 |
+
"eval_steps_per_second": 1.454,
|
| 1025 |
+
"step": 31500
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.1219802952210651,
|
| 1029 |
+
"grad_norm": 4.273074626922607,
|
| 1030 |
+
"learning_rate": 1.9919181977348992e-05,
|
| 1031 |
+
"loss": 9.4714,
|
| 1032 |
+
"step": 32000
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.1219802952210651,
|
| 1036 |
+
"eval_cosine_accuracy": 0.9416828751564026,
|
| 1037 |
+
"eval_loss": 9.125110626220703,
|
| 1038 |
+
"eval_runtime": 50.5517,
|
| 1039 |
+
"eval_samples_per_second": 374.191,
|
| 1040 |
+
"eval_steps_per_second": 1.464,
|
| 1041 |
+
"step": 32000
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 1.1395112373338943,
|
| 1045 |
+
"grad_norm": 3.9410202503204346,
|
| 1046 |
+
"learning_rate": 1.989421920691003e-05,
|
| 1047 |
+
"loss": 9.474,
|
| 1048 |
+
"step": 32500
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 1.1395112373338943,
|
| 1052 |
+
"eval_cosine_accuracy": 0.9401875734329224,
|
| 1053 |
+
"eval_loss": 9.20176887512207,
|
| 1054 |
+
"eval_runtime": 51.617,
|
| 1055 |
+
"eval_samples_per_second": 366.468,
|
| 1056 |
+
"eval_steps_per_second": 1.434,
|
| 1057 |
+
"step": 32500
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 1.1570421794467234,
|
| 1061 |
+
"grad_norm": 5.124668598175049,
|
| 1062 |
+
"learning_rate": 1.9865921965609503e-05,
|
| 1063 |
+
"loss": 9.4644,
|
| 1064 |
+
"step": 33000
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 1.1570421794467234,
|
| 1068 |
+
"eval_cosine_accuracy": 0.9374235272407532,
|
| 1069 |
+
"eval_loss": 9.209524154663086,
|
| 1070 |
+
"eval_runtime": 50.948,
|
| 1071 |
+
"eval_samples_per_second": 371.281,
|
| 1072 |
+
"eval_steps_per_second": 1.452,
|
| 1073 |
+
"step": 33000
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 1.1745731215595527,
|
| 1077 |
+
"grad_norm": 8.401219367980957,
|
| 1078 |
+
"learning_rate": 1.9834366345696648e-05,
|
| 1079 |
+
"loss": 9.4593,
|
| 1080 |
+
"step": 33500
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 1.1745731215595527,
|
| 1084 |
+
"eval_cosine_accuracy": 0.9377407431602478,
|
| 1085 |
+
"eval_loss": 9.227911949157715,
|
| 1086 |
+
"eval_runtime": 50.5886,
|
| 1087 |
+
"eval_samples_per_second": 373.918,
|
| 1088 |
+
"eval_steps_per_second": 1.463,
|
| 1089 |
+
"step": 33500
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 1.1921040636723816,
|
| 1093 |
+
"grad_norm": 3.5143702030181885,
|
| 1094 |
+
"learning_rate": 1.9799436509923552e-05,
|
| 1095 |
+
"loss": 9.4608,
|
| 1096 |
+
"step": 34000
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 1.1921040636723816,
|
| 1100 |
+
"eval_cosine_accuracy": 0.9348860383033752,
|
| 1101 |
+
"eval_loss": 9.227869033813477,
|
| 1102 |
+
"eval_runtime": 51.2403,
|
| 1103 |
+
"eval_samples_per_second": 369.163,
|
| 1104 |
+
"eval_steps_per_second": 1.444,
|
| 1105 |
+
"step": 34000
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 1.209635005785211,
|
| 1109 |
+
"grad_norm": 4.010260581970215,
|
| 1110 |
+
"learning_rate": 1.9761204146197647e-05,
|
| 1111 |
+
"loss": 9.451,
|
| 1112 |
+
"step": 34500
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 1.209635005785211,
|
| 1116 |
+
"eval_cosine_accuracy": 0.9332094788551331,
|
| 1117 |
+
"eval_loss": 9.30997085571289,
|
| 1118 |
+
"eval_runtime": 50.2953,
|
| 1119 |
+
"eval_samples_per_second": 376.099,
|
| 1120 |
+
"eval_steps_per_second": 1.471,
|
| 1121 |
+
"step": 34500
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 1.2271659478980401,
|
| 1125 |
+
"grad_norm": 4.186135768890381,
|
| 1126 |
+
"learning_rate": 1.9719682139285293e-05,
|
| 1127 |
+
"loss": 9.4515,
|
| 1128 |
+
"step": 35000
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 1.2271659478980401,
|
| 1132 |
+
"eval_cosine_accuracy": 0.9390094876289368,
|
| 1133 |
+
"eval_loss": 9.19826889038086,
|
| 1134 |
+
"eval_runtime": 50.7328,
|
| 1135 |
+
"eval_samples_per_second": 372.855,
|
| 1136 |
+
"eval_steps_per_second": 1.459,
|
| 1137 |
+
"step": 35000
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.2446968900108693,
|
| 1141 |
+
"grad_norm": 3.7727272510528564,
|
| 1142 |
+
"learning_rate": 1.967488448260217e-05,
|
| 1143 |
+
"loss": 9.4507,
|
| 1144 |
+
"step": 35500
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.2446968900108693,
|
| 1148 |
+
"eval_cosine_accuracy": 0.9352485537528992,
|
| 1149 |
+
"eval_loss": 9.263128280639648,
|
| 1150 |
+
"eval_runtime": 49.9694,
|
| 1151 |
+
"eval_samples_per_second": 378.552,
|
| 1152 |
+
"eval_steps_per_second": 1.481,
|
| 1153 |
+
"step": 35500
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 1.2622278321236984,
|
| 1157 |
+
"grad_norm": 4.541546821594238,
|
| 1158 |
+
"learning_rate": 1.9626826273497317e-05,
|
| 1159 |
+
"loss": 9.4483,
|
| 1160 |
+
"step": 36000
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 1.2622278321236984,
|
| 1164 |
+
"eval_cosine_accuracy": 0.9351125955581665,
|
| 1165 |
+
"eval_loss": 9.285823822021484,
|
| 1166 |
+
"eval_runtime": 50.176,
|
| 1167 |
+
"eval_samples_per_second": 376.993,
|
| 1168 |
+
"eval_steps_per_second": 1.475,
|
| 1169 |
+
"step": 36000
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 1.2797587742365275,
|
| 1173 |
+
"grad_norm": 5.939820289611816,
|
| 1174 |
+
"learning_rate": 1.957562953985298e-05,
|
| 1175 |
+
"loss": 9.4421,
|
| 1176 |
+
"step": 36500
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 1.2797587742365275,
|
| 1180 |
+
"eval_cosine_accuracy": 0.9381938576698303,
|
| 1181 |
+
"eval_loss": 9.222258567810059,
|
| 1182 |
+
"eval_runtime": 50.9164,
|
| 1183 |
+
"eval_samples_per_second": 371.511,
|
| 1184 |
+
"eval_steps_per_second": 1.453,
|
| 1185 |
+
"step": 36500
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 1.2972897163493566,
|
| 1189 |
+
"grad_norm": 6.257410526275635,
|
| 1190 |
+
"learning_rate": 1.9521106344029423e-05,
|
| 1191 |
+
"loss": 9.4373,
|
| 1192 |
+
"step": 37000
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 1.2972897163493566,
|
| 1196 |
+
"eval_cosine_accuracy": 0.9351125955581665,
|
| 1197 |
+
"eval_loss": 9.2606840133667,
|
| 1198 |
+
"eval_runtime": 49.7217,
|
| 1199 |
+
"eval_samples_per_second": 380.437,
|
| 1200 |
+
"eval_steps_per_second": 1.488,
|
| 1201 |
+
"step": 37000
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 1.3148206584621858,
|
| 1205 |
+
"grad_norm": 4.259251117706299,
|
| 1206 |
+
"learning_rate": 1.9463374420866178e-05,
|
| 1207 |
+
"loss": 9.4349,
|
| 1208 |
+
"step": 37500
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 1.3148206584621858,
|
| 1212 |
+
"eval_cosine_accuracy": 0.934478223323822,
|
| 1213 |
+
"eval_loss": 9.276906967163086,
|
| 1214 |
+
"eval_runtime": 51.3387,
|
| 1215 |
+
"eval_samples_per_second": 368.455,
|
| 1216 |
+
"eval_steps_per_second": 1.441,
|
| 1217 |
+
"step": 37500
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 1.332351600575015,
|
| 1221 |
+
"grad_norm": 6.372994899749756,
|
| 1222 |
+
"learning_rate": 1.940245322671565e-05,
|
| 1223 |
+
"loss": 9.4263,
|
| 1224 |
+
"step": 38000
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 1.332351600575015,
|
| 1228 |
+
"eval_cosine_accuracy": 0.9330282211303711,
|
| 1229 |
+
"eval_loss": 9.382668495178223,
|
| 1230 |
+
"eval_runtime": 51.789,
|
| 1231 |
+
"eval_samples_per_second": 365.252,
|
| 1232 |
+
"eval_steps_per_second": 1.429,
|
| 1233 |
+
"step": 38000
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 1.349882542687844,
|
| 1237 |
+
"grad_norm": 4.699613094329834,
|
| 1238 |
+
"learning_rate": 1.933836329275292e-05,
|
| 1239 |
+
"loss": 9.4243,
|
| 1240 |
+
"step": 38500
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 1.349882542687844,
|
| 1244 |
+
"eval_cosine_accuracy": 0.9343876242637634,
|
| 1245 |
+
"eval_loss": 9.369087219238281,
|
| 1246 |
+
"eval_runtime": 49.8777,
|
| 1247 |
+
"eval_samples_per_second": 379.247,
|
| 1248 |
+
"eval_steps_per_second": 1.484,
|
| 1249 |
+
"step": 38500
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.3674134848006732,
|
| 1253 |
+
"grad_norm": 4.506259918212891,
|
| 1254 |
+
"learning_rate": 1.9271263818166978e-05,
|
| 1255 |
+
"loss": 9.4281,
|
| 1256 |
+
"step": 39000
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.3674134848006732,
|
| 1260 |
+
"eval_cosine_accuracy": 0.9313516616821289,
|
| 1261 |
+
"eval_loss": 9.409356117248535,
|
| 1262 |
+
"eval_runtime": 50.7091,
|
| 1263 |
+
"eval_samples_per_second": 373.03,
|
| 1264 |
+
"eval_steps_per_second": 1.459,
|
| 1265 |
+
"step": 39000
|
| 1266 |
+
},
|
| 1267 |
+
{
|
| 1268 |
+
"epoch": 1.3849444269135023,
|
| 1269 |
+
"grad_norm": 4.799638271331787,
|
| 1270 |
+
"learning_rate": 1.9200908488086076e-05,
|
| 1271 |
+
"loss": 9.4189,
|
| 1272 |
+
"step": 39500
|
| 1273 |
+
},
|
| 1274 |
+
{
|
| 1275 |
+
"epoch": 1.3849444269135023,
|
| 1276 |
+
"eval_cosine_accuracy": 0.9301282167434692,
|
| 1277 |
+
"eval_loss": 9.410999298095703,
|
| 1278 |
+
"eval_runtime": 49.7029,
|
| 1279 |
+
"eval_samples_per_second": 380.581,
|
| 1280 |
+
"eval_steps_per_second": 1.489,
|
| 1281 |
+
"step": 39500
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 1.4024753690263314,
|
| 1285 |
+
"grad_norm": 4.841325759887695,
|
| 1286 |
+
"learning_rate": 1.912745234119368e-05,
|
| 1287 |
+
"loss": 9.4184,
|
| 1288 |
+
"step": 40000
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"epoch": 1.4024753690263314,
|
| 1292 |
+
"eval_cosine_accuracy": 0.9344329237937927,
|
| 1293 |
+
"eval_loss": 9.38805866241455,
|
| 1294 |
+
"eval_runtime": 50.6935,
|
| 1295 |
+
"eval_samples_per_second": 373.144,
|
| 1296 |
+
"eval_steps_per_second": 1.46,
|
| 1297 |
+
"step": 40000
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 1.4200063111391605,
|
| 1301 |
+
"grad_norm": 5.196623802185059,
|
| 1302 |
+
"learning_rate": 1.9050920133094675e-05,
|
| 1303 |
+
"loss": 9.4197,
|
| 1304 |
+
"step": 40500
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 1.4200063111391605,
|
| 1308 |
+
"eval_cosine_accuracy": 0.9307172894477844,
|
| 1309 |
+
"eval_loss": 9.490105628967285,
|
| 1310 |
+
"eval_runtime": 50.1087,
|
| 1311 |
+
"eval_samples_per_second": 377.5,
|
| 1312 |
+
"eval_steps_per_second": 1.477,
|
| 1313 |
+
"step": 40500
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 1.4375372532519899,
|
| 1317 |
+
"grad_norm": 5.746089935302734,
|
| 1318 |
+
"learning_rate": 1.8971337656063586e-05,
|
| 1319 |
+
"loss": 9.4085,
|
| 1320 |
+
"step": 41000
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 1.4375372532519899,
|
| 1324 |
+
"eval_cosine_accuracy": 0.9347047805786133,
|
| 1325 |
+
"eval_loss": 9.269803047180176,
|
| 1326 |
+
"eval_runtime": 52.0188,
|
| 1327 |
+
"eval_samples_per_second": 363.638,
|
| 1328 |
+
"eval_steps_per_second": 1.423,
|
| 1329 |
+
"step": 41000
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 1.4550681953648188,
|
| 1333 |
+
"grad_norm": 4.658451557159424,
|
| 1334 |
+
"learning_rate": 1.8888731730352233e-05,
|
| 1335 |
+
"loss": 9.4137,
|
| 1336 |
+
"step": 41500
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 1.4550681953648188,
|
| 1340 |
+
"eval_cosine_accuracy": 0.9356563687324524,
|
| 1341 |
+
"eval_loss": 9.31067180633545,
|
| 1342 |
+
"eval_runtime": 51.6789,
|
| 1343 |
+
"eval_samples_per_second": 366.03,
|
| 1344 |
+
"eval_steps_per_second": 1.432,
|
| 1345 |
+
"step": 41500
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 1.4725991374776481,
|
| 1349 |
+
"grad_norm": 5.862908363342285,
|
| 1350 |
+
"learning_rate": 1.8803130195151012e-05,
|
| 1351 |
+
"loss": 9.4056,
|
| 1352 |
+
"step": 42000
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 1.4725991374776481,
|
| 1356 |
+
"eval_cosine_accuracy": 0.9329829216003418,
|
| 1357 |
+
"eval_loss": 9.364194869995117,
|
| 1358 |
+
"eval_runtime": 51.0292,
|
| 1359 |
+
"eval_samples_per_second": 370.69,
|
| 1360 |
+
"eval_steps_per_second": 1.45,
|
| 1361 |
+
"step": 42000
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.4901300795904773,
|
| 1365 |
+
"grad_norm": 3.8584418296813965,
|
| 1366 |
+
"learning_rate": 1.8714741976951926e-05,
|
| 1367 |
+
"loss": 9.4021,
|
| 1368 |
+
"step": 42500
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.4901300795904773,
|
| 1372 |
+
"eval_cosine_accuracy": 0.9347047805786133,
|
| 1373 |
+
"eval_loss": 9.337225914001465,
|
| 1374 |
+
"eval_runtime": 50.3554,
|
| 1375 |
+
"eval_samples_per_second": 375.65,
|
| 1376 |
+
"eval_steps_per_second": 1.47,
|
| 1377 |
+
"step": 42500
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 1.5076610217033064,
|
| 1381 |
+
"grad_norm": 6.5607805252075195,
|
| 1382 |
+
"learning_rate": 1.8623242612221316e-05,
|
| 1383 |
+
"loss": 9.4044,
|
| 1384 |
+
"step": 43000
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 1.5076610217033064,
|
| 1388 |
+
"eval_cosine_accuracy": 0.9298110604286194,
|
| 1389 |
+
"eval_loss": 9.429045677185059,
|
| 1390 |
+
"eval_runtime": 50.7892,
|
| 1391 |
+
"eval_samples_per_second": 372.442,
|
| 1392 |
+
"eval_steps_per_second": 1.457,
|
| 1393 |
+
"step": 43000
|
| 1394 |
+
},
|
| 1395 |
+
{
|
| 1396 |
+
"epoch": 1.5251919638161355,
|
| 1397 |
+
"grad_norm": 4.057403087615967,
|
| 1398 |
+
"learning_rate": 1.8528837111026442e-05,
|
| 1399 |
+
"loss": 9.4015,
|
| 1400 |
+
"step": 43500
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 1.5251919638161355,
|
| 1404 |
+
"eval_cosine_accuracy": 0.9316688776016235,
|
| 1405 |
+
"eval_loss": 9.397970199584961,
|
| 1406 |
+
"eval_runtime": 50.1715,
|
| 1407 |
+
"eval_samples_per_second": 377.027,
|
| 1408 |
+
"eval_steps_per_second": 1.475,
|
| 1409 |
+
"step": 43500
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 1.5427229059289647,
|
| 1413 |
+
"grad_norm": 6.545431613922119,
|
| 1414 |
+
"learning_rate": 1.8431557289156537e-05,
|
| 1415 |
+
"loss": 9.3987,
|
| 1416 |
+
"step": 44000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 1.5427229059289647,
|
| 1420 |
+
"eval_cosine_accuracy": 0.9302188754081726,
|
| 1421 |
+
"eval_loss": 9.482829093933105,
|
| 1422 |
+
"eval_runtime": 49.9881,
|
| 1423 |
+
"eval_samples_per_second": 378.41,
|
| 1424 |
+
"eval_steps_per_second": 1.48,
|
| 1425 |
+
"step": 44000
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 1.5602538480417938,
|
| 1429 |
+
"grad_norm": 4.163936614990234,
|
| 1430 |
+
"learning_rate": 1.8331638987381773e-05,
|
| 1431 |
+
"loss": 9.3933,
|
| 1432 |
+
"step": 44500
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 1.5602538480417938,
|
| 1436 |
+
"eval_cosine_accuracy": 0.928632915019989,
|
| 1437 |
+
"eval_loss": 9.449180603027344,
|
| 1438 |
+
"eval_runtime": 51.351,
|
| 1439 |
+
"eval_samples_per_second": 368.367,
|
| 1440 |
+
"eval_steps_per_second": 1.441,
|
| 1441 |
+
"step": 44500
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 1.577784790154623,
|
| 1445 |
+
"grad_norm": 5.566523551940918,
|
| 1446 |
+
"learning_rate": 1.8228715416485997e-05,
|
| 1447 |
+
"loss": 9.3927,
|
| 1448 |
+
"step": 45000
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 1.577784790154623,
|
| 1452 |
+
"eval_cosine_accuracy": 0.9261860251426697,
|
| 1453 |
+
"eval_loss": 9.525978088378906,
|
| 1454 |
+
"eval_runtime": 50.7927,
|
| 1455 |
+
"eval_samples_per_second": 372.416,
|
| 1456 |
+
"eval_steps_per_second": 1.457,
|
| 1457 |
+
"step": 45000
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 1.595315732267452,
|
| 1461 |
+
"grad_norm": 5.857015132904053,
|
| 1462 |
+
"learning_rate": 1.8123018669535538e-05,
|
| 1463 |
+
"loss": 9.3897,
|
| 1464 |
+
"step": 45500
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 1.595315732267452,
|
| 1468 |
+
"eval_cosine_accuracy": 0.9293125867843628,
|
| 1469 |
+
"eval_loss": 9.421846389770508,
|
| 1470 |
+
"eval_runtime": 50.3281,
|
| 1471 |
+
"eval_samples_per_second": 375.854,
|
| 1472 |
+
"eval_steps_per_second": 1.47,
|
| 1473 |
+
"step": 45500
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.6128466743802812,
|
| 1477 |
+
"grad_norm": 3.6868417263031006,
|
| 1478 |
+
"learning_rate": 1.8014584367605247e-05,
|
| 1479 |
+
"loss": 9.3906,
|
| 1480 |
+
"step": 46000
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.6128466743802812,
|
| 1484 |
+
"eval_cosine_accuracy": 0.9294938445091248,
|
| 1485 |
+
"eval_loss": 9.40963077545166,
|
| 1486 |
+
"eval_runtime": 50.7085,
|
| 1487 |
+
"eval_samples_per_second": 373.034,
|
| 1488 |
+
"eval_steps_per_second": 1.459,
|
| 1489 |
+
"step": 46000
|
| 1490 |
+
},
|
| 1491 |
+
{
|
| 1492 |
+
"epoch": 1.6303776164931103,
|
| 1493 |
+
"grad_norm": 3.6633148193359375,
|
| 1494 |
+
"learning_rate": 1.7903449054358905e-05,
|
| 1495 |
+
"loss": 9.3808,
|
| 1496 |
+
"step": 46500
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 1.6303776164931103,
|
| 1500 |
+
"eval_cosine_accuracy": 0.926276683807373,
|
| 1501 |
+
"eval_loss": 9.574146270751953,
|
| 1502 |
+
"eval_runtime": 49.9467,
|
| 1503 |
+
"eval_samples_per_second": 378.723,
|
| 1504 |
+
"eval_steps_per_second": 1.482,
|
| 1505 |
+
"step": 46500
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 1.6479085586059394,
|
| 1509 |
+
"grad_norm": 4.91319465637207,
|
| 1510 |
+
"learning_rate": 1.778988041436223e-05,
|
| 1511 |
+
"loss": 9.3767,
|
| 1512 |
+
"step": 47000
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 1.6479085586059394,
|
| 1516 |
+
"eval_cosine_accuracy": 0.9274548292160034,
|
| 1517 |
+
"eval_loss": 9.482501029968262,
|
| 1518 |
+
"eval_runtime": 50.8995,
|
| 1519 |
+
"eval_samples_per_second": 371.634,
|
| 1520 |
+
"eval_steps_per_second": 1.454,
|
| 1521 |
+
"step": 47000
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 1.6654395007187688,
|
| 1525 |
+
"grad_norm": 3.4284417629241943,
|
| 1526 |
+
"learning_rate": 1.7673461549492397e-05,
|
| 1527 |
+
"loss": 9.3764,
|
| 1528 |
+
"step": 47500
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 1.6654395007187688,
|
| 1532 |
+
"eval_cosine_accuracy": 0.9308985471725464,
|
| 1533 |
+
"eval_loss": 9.456860542297363,
|
| 1534 |
+
"eval_runtime": 50.2147,
|
| 1535 |
+
"eval_samples_per_second": 376.703,
|
| 1536 |
+
"eval_steps_per_second": 1.474,
|
| 1537 |
+
"step": 47500
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 1.6829704428315977,
|
| 1541 |
+
"grad_norm": 4.275540351867676,
|
| 1542 |
+
"learning_rate": 1.75544566357986e-05,
|
| 1543 |
+
"loss": 9.3764,
|
| 1544 |
+
"step": 48000
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 1.6829704428315977,
|
| 1548 |
+
"eval_cosine_accuracy": 0.9290407299995422,
|
| 1549 |
+
"eval_loss": 9.518174171447754,
|
| 1550 |
+
"eval_runtime": 50.3276,
|
| 1551 |
+
"eval_samples_per_second": 375.857,
|
| 1552 |
+
"eval_steps_per_second": 1.47,
|
| 1553 |
+
"step": 48000
|
| 1554 |
+
},
|
| 1555 |
+
{
|
| 1556 |
+
"epoch": 1.700501384944427,
|
| 1557 |
+
"grad_norm": 5.003416061401367,
|
| 1558 |
+
"learning_rate": 1.7432905779367918e-05,
|
| 1559 |
+
"loss": 9.3742,
|
| 1560 |
+
"step": 48500
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 1.700501384944427,
|
| 1564 |
+
"eval_cosine_accuracy": 0.9265032410621643,
|
| 1565 |
+
"eval_loss": 9.55920696258545,
|
| 1566 |
+
"eval_runtime": 51.3499,
|
| 1567 |
+
"eval_samples_per_second": 368.375,
|
| 1568 |
+
"eval_steps_per_second": 1.441,
|
| 1569 |
+
"step": 48500
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 1.718032327057256,
|
| 1573 |
+
"grad_norm": 4.121612071990967,
|
| 1574 |
+
"learning_rate": 1.7309100528290082e-05,
|
| 1575 |
+
"loss": 9.376,
|
| 1576 |
+
"step": 49000
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 1.718032327057256,
|
| 1580 |
+
"eval_cosine_accuracy": 0.9295845031738281,
|
| 1581 |
+
"eval_loss": 9.498641967773438,
|
| 1582 |
+
"eval_runtime": 50.6942,
|
| 1583 |
+
"eval_samples_per_second": 373.139,
|
| 1584 |
+
"eval_steps_per_second": 1.46,
|
| 1585 |
+
"step": 49000
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.7355632691700853,
|
| 1589 |
+
"grad_norm": 4.586385726928711,
|
| 1590 |
+
"learning_rate": 1.718258640695659e-05,
|
| 1591 |
+
"loss": 9.3663,
|
| 1592 |
+
"step": 49500
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.7355632691700853,
|
| 1596 |
+
"eval_cosine_accuracy": 0.9289047718048096,
|
| 1597 |
+
"eval_loss": 9.542733192443848,
|
| 1598 |
+
"eval_runtime": 50.1591,
|
| 1599 |
+
"eval_samples_per_second": 377.12,
|
| 1600 |
+
"eval_steps_per_second": 1.475,
|
| 1601 |
+
"step": 49500
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"epoch": 1.7530942112829142,
|
| 1605 |
+
"grad_norm": 4.653346061706543,
|
| 1606 |
+
"learning_rate": 1.7053651667625475e-05,
|
| 1607 |
+
"loss": 9.3676,
|
| 1608 |
+
"step": 50000
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 1.7530942112829142,
|
| 1612 |
+
"eval_cosine_accuracy": 0.9276360273361206,
|
| 1613 |
+
"eval_loss": 9.607583999633789,
|
| 1614 |
+
"eval_runtime": 51.1317,
|
| 1615 |
+
"eval_samples_per_second": 369.946,
|
| 1616 |
+
"eval_steps_per_second": 1.447,
|
| 1617 |
+
"step": 50000
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 1.7706251533957436,
|
| 1621 |
+
"grad_norm": 7.506467819213867,
|
| 1622 |
+
"learning_rate": 1.692233976285448e-05,
|
| 1623 |
+
"loss": 9.3654,
|
| 1624 |
+
"step": 50500
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"epoch": 1.7706251533957436,
|
| 1628 |
+
"eval_cosine_accuracy": 0.9270923137664795,
|
| 1629 |
+
"eval_loss": 9.642268180847168,
|
| 1630 |
+
"eval_runtime": 50.1614,
|
| 1631 |
+
"eval_samples_per_second": 377.103,
|
| 1632 |
+
"eval_steps_per_second": 1.475,
|
| 1633 |
+
"step": 50500
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 1.7881560955085727,
|
| 1637 |
+
"grad_norm": 5.075575828552246,
|
| 1638 |
+
"learning_rate": 1.678896453440646e-05,
|
| 1639 |
+
"loss": 9.3648,
|
| 1640 |
+
"step": 51000
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 1.7881560955085727,
|
| 1644 |
+
"eval_cosine_accuracy": 0.928814172744751,
|
| 1645 |
+
"eval_loss": 9.493197441101074,
|
| 1646 |
+
"eval_runtime": 50.9437,
|
| 1647 |
+
"eval_samples_per_second": 371.312,
|
| 1648 |
+
"eval_steps_per_second": 1.453,
|
| 1649 |
+
"step": 51000
|
| 1650 |
+
},
|
| 1651 |
+
{
|
| 1652 |
+
"epoch": 1.8056870376214018,
|
| 1653 |
+
"grad_norm": 4.172321319580078,
|
| 1654 |
+
"learning_rate": 1.665303637632705e-05,
|
| 1655 |
+
"loss": 9.3683,
|
| 1656 |
+
"step": 51500
|
| 1657 |
+
},
|
| 1658 |
+
{
|
| 1659 |
+
"epoch": 1.8056870376214018,
|
| 1660 |
+
"eval_cosine_accuracy": 0.9274094700813293,
|
| 1661 |
+
"eval_loss": 9.589746475219727,
|
| 1662 |
+
"eval_runtime": 50.257,
|
| 1663 |
+
"eval_samples_per_second": 376.386,
|
| 1664 |
+
"eval_steps_per_second": 1.472,
|
| 1665 |
+
"step": 51500
|
| 1666 |
+
},
|
| 1667 |
+
{
|
| 1668 |
+
"epoch": 1.823217979734231,
|
| 1669 |
+
"grad_norm": 5.439345359802246,
|
| 1670 |
+
"learning_rate": 1.6514866064978768e-05,
|
| 1671 |
+
"loss": 9.3553,
|
| 1672 |
+
"step": 52000
|
| 1673 |
+
},
|
| 1674 |
+
{
|
| 1675 |
+
"epoch": 1.823217979734231,
|
| 1676 |
+
"eval_cosine_accuracy": 0.9257329106330872,
|
| 1677 |
+
"eval_loss": 9.655684471130371,
|
| 1678 |
+
"eval_runtime": 50.1972,
|
| 1679 |
+
"eval_samples_per_second": 376.834,
|
| 1680 |
+
"eval_steps_per_second": 1.474,
|
| 1681 |
+
"step": 52000
|
| 1682 |
+
},
|
| 1683 |
+
{
|
| 1684 |
+
"epoch": 1.84074892184706,
|
| 1685 |
+
"grad_norm": 5.535743236541748,
|
| 1686 |
+
"learning_rate": 1.6374500165418183e-05,
|
| 1687 |
+
"loss": 9.3539,
|
| 1688 |
+
"step": 52500
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 1.84074892184706,
|
| 1692 |
+
"eval_cosine_accuracy": 0.9289501309394836,
|
| 1693 |
+
"eval_loss": 9.585236549377441,
|
| 1694 |
+
"eval_runtime": 51.4254,
|
| 1695 |
+
"eval_samples_per_second": 367.834,
|
| 1696 |
+
"eval_steps_per_second": 1.439,
|
| 1697 |
+
"step": 52500
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 1.8582798639598892,
|
| 1701 |
+
"grad_norm": 4.505749225616455,
|
| 1702 |
+
"learning_rate": 1.6231985982641494e-05,
|
| 1703 |
+
"loss": 9.3533,
|
| 1704 |
+
"step": 53000
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 1.8582798639598892,
|
| 1708 |
+
"eval_cosine_accuracy": 0.9288594722747803,
|
| 1709 |
+
"eval_loss": 9.569564819335938,
|
| 1710 |
+
"eval_runtime": 50.8164,
|
| 1711 |
+
"eval_samples_per_second": 372.242,
|
| 1712 |
+
"eval_steps_per_second": 1.456,
|
| 1713 |
+
"step": 53000
|
| 1714 |
+
},
|
| 1715 |
+
{
|
| 1716 |
+
"epoch": 1.8758108060727183,
|
| 1717 |
+
"grad_norm": 6.0967535972595215,
|
| 1718 |
+
"learning_rate": 1.6087662838292334e-05,
|
| 1719 |
+
"loss": 9.3567,
|
| 1720 |
+
"step": 53500
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 1.8758108060727183,
|
| 1724 |
+
"eval_cosine_accuracy": 0.9304001331329346,
|
| 1725 |
+
"eval_loss": 9.577661514282227,
|
| 1726 |
+
"eval_runtime": 50.034,
|
| 1727 |
+
"eval_samples_per_second": 378.063,
|
| 1728 |
+
"eval_steps_per_second": 1.479,
|
| 1729 |
+
"step": 53500
|
| 1730 |
+
},
|
| 1731 |
+
{
|
| 1732 |
+
"epoch": 1.8933417481855475,
|
| 1733 |
+
"grad_norm": 3.9358584880828857,
|
| 1734 |
+
"learning_rate": 1.594100093780804e-05,
|
| 1735 |
+
"loss": 9.3531,
|
| 1736 |
+
"step": 54000
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 1.8933417481855475,
|
| 1740 |
+
"eval_cosine_accuracy": 0.9263219833374023,
|
| 1741 |
+
"eval_loss": 9.534945487976074,
|
| 1742 |
+
"eval_runtime": 50.8761,
|
| 1743 |
+
"eval_samples_per_second": 371.805,
|
| 1744 |
+
"eval_steps_per_second": 1.455,
|
| 1745 |
+
"step": 54000
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 1.9108726902983766,
|
| 1749 |
+
"grad_norm": 3.7857773303985596,
|
| 1750 |
+
"learning_rate": 1.5792336848560662e-05,
|
| 1751 |
+
"loss": 9.3452,
|
| 1752 |
+
"step": 54500
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 1.9108726902983766,
|
| 1756 |
+
"eval_cosine_accuracy": 0.9293125867843628,
|
| 1757 |
+
"eval_loss": 9.485892295837402,
|
| 1758 |
+
"eval_runtime": 50.8296,
|
| 1759 |
+
"eval_samples_per_second": 372.145,
|
| 1760 |
+
"eval_steps_per_second": 1.456,
|
| 1761 |
+
"step": 54500
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 1.928403632411206,
|
| 1765 |
+
"grad_norm": 4.47232723236084,
|
| 1766 |
+
"learning_rate": 1.564172067213612e-05,
|
| 1767 |
+
"loss": 9.3467,
|
| 1768 |
+
"step": 55000
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"epoch": 1.928403632411206,
|
| 1772 |
+
"eval_cosine_accuracy": 0.9282703995704651,
|
| 1773 |
+
"eval_loss": 9.556325912475586,
|
| 1774 |
+
"eval_runtime": 51.0784,
|
| 1775 |
+
"eval_samples_per_second": 370.332,
|
| 1776 |
+
"eval_steps_per_second": 1.449,
|
| 1777 |
+
"step": 55000
|
| 1778 |
+
},
|
| 1779 |
+
{
|
| 1780 |
+
"epoch": 1.9459345745240348,
|
| 1781 |
+
"grad_norm": 5.011724948883057,
|
| 1782 |
+
"learning_rate": 1.548920316799719e-05,
|
| 1783 |
+
"loss": 9.345,
|
| 1784 |
+
"step": 55500
|
| 1785 |
+
},
|
| 1786 |
+
{
|
| 1787 |
+
"epoch": 1.9459345745240348,
|
| 1788 |
+
"eval_cosine_accuracy": 0.9248266816139221,
|
| 1789 |
+
"eval_loss": 9.641026496887207,
|
| 1790 |
+
"eval_runtime": 49.7808,
|
| 1791 |
+
"eval_samples_per_second": 379.986,
|
| 1792 |
+
"eval_steps_per_second": 1.487,
|
| 1793 |
+
"step": 55500
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 1.9634655166368642,
|
| 1797 |
+
"grad_norm": 5.616572856903076,
|
| 1798 |
+
"learning_rate": 1.5335146282991408e-05,
|
| 1799 |
+
"loss": 9.3447,
|
| 1800 |
+
"step": 56000
|
| 1801 |
+
},
|
| 1802 |
+
{
|
| 1803 |
+
"epoch": 1.9634655166368642,
|
| 1804 |
+
"eval_cosine_accuracy": 0.924101710319519,
|
| 1805 |
+
"eval_loss": 9.674759864807129,
|
| 1806 |
+
"eval_runtime": 51.3719,
|
| 1807 |
+
"eval_samples_per_second": 368.217,
|
| 1808 |
+
"eval_steps_per_second": 1.44,
|
| 1809 |
+
"step": 56000
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 1.980996458749693,
|
| 1813 |
+
"grad_norm": 4.511038303375244,
|
| 1814 |
+
"learning_rate": 1.5178984491053066e-05,
|
| 1815 |
+
"loss": 9.3448,
|
| 1816 |
+
"step": 56500
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 1.980996458749693,
|
| 1820 |
+
"eval_cosine_accuracy": 0.9255063533782959,
|
| 1821 |
+
"eval_loss": 9.623557090759277,
|
| 1822 |
+
"eval_runtime": 50.1837,
|
| 1823 |
+
"eval_samples_per_second": 376.935,
|
| 1824 |
+
"eval_steps_per_second": 1.475,
|
| 1825 |
+
"step": 56500
|
| 1826 |
+
},
|
| 1827 |
+
{
|
| 1828 |
+
"epoch": 1.9985274008625225,
|
| 1829 |
+
"grad_norm": 4.871715545654297,
|
| 1830 |
+
"learning_rate": 1.5021077319058275e-05,
|
| 1831 |
+
"loss": 9.3432,
|
| 1832 |
+
"step": 57000
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"epoch": 1.9985274008625225,
|
| 1836 |
+
"eval_cosine_accuracy": 0.9239204525947571,
|
| 1837 |
+
"eval_loss": 9.694854736328125,
|
| 1838 |
+
"eval_runtime": 49.8583,
|
| 1839 |
+
"eval_samples_per_second": 379.395,
|
| 1840 |
+
"eval_steps_per_second": 1.484,
|
| 1841 |
+
"step": 57000
|
| 1842 |
+
},
|
| 1843 |
+
{
|
| 1844 |
+
"epoch": 2.0160583429753514,
|
| 1845 |
+
"grad_norm": 5.141265869140625,
|
| 1846 |
+
"learning_rate": 1.4861477983623002e-05,
|
| 1847 |
+
"loss": 9.317,
|
| 1848 |
+
"step": 57500
|
| 1849 |
+
},
|
| 1850 |
+
{
|
| 1851 |
+
"epoch": 2.0160583429753514,
|
| 1852 |
+
"eval_cosine_accuracy": 0.9253250956535339,
|
| 1853 |
+
"eval_loss": 9.723149299621582,
|
| 1854 |
+
"eval_runtime": 50.877,
|
| 1855 |
+
"eval_samples_per_second": 371.799,
|
| 1856 |
+
"eval_steps_per_second": 1.454,
|
| 1857 |
+
"step": 57500
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 2.0335892850881807,
|
| 1861 |
+
"grad_norm": 6.244866847991943,
|
| 1862 |
+
"learning_rate": 1.4700564346139085e-05,
|
| 1863 |
+
"loss": 9.3136,
|
| 1864 |
+
"step": 58000
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 2.0335892850881807,
|
| 1868 |
+
"eval_cosine_accuracy": 0.9257782697677612,
|
| 1869 |
+
"eval_loss": 9.7728853225708,
|
| 1870 |
+
"eval_runtime": 50.1884,
|
| 1871 |
+
"eval_samples_per_second": 376.9,
|
| 1872 |
+
"eval_steps_per_second": 1.474,
|
| 1873 |
+
"step": 58000
|
| 1874 |
+
},
|
| 1875 |
+
{
|
| 1876 |
+
"epoch": 2.0511202272010096,
|
| 1877 |
+
"grad_norm": 4.268733978271484,
|
| 1878 |
+
"learning_rate": 1.4537745710148372e-05,
|
| 1879 |
+
"loss": 9.3189,
|
| 1880 |
+
"step": 58500
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 2.0511202272010096,
|
| 1884 |
+
"eval_cosine_accuracy": 0.926457941532135,
|
| 1885 |
+
"eval_loss": 9.745859146118164,
|
| 1886 |
+
"eval_runtime": 50.6296,
|
| 1887 |
+
"eval_samples_per_second": 373.615,
|
| 1888 |
+
"eval_steps_per_second": 1.462,
|
| 1889 |
+
"step": 58500
|
| 1890 |
+
},
|
| 1891 |
+
{
|
| 1892 |
+
"epoch": 2.068651169313839,
|
| 1893 |
+
"grad_norm": 5.612603664398193,
|
| 1894 |
+
"learning_rate": 1.4373397799281979e-05,
|
| 1895 |
+
"loss": 9.3126,
|
| 1896 |
+
"step": 59000
|
| 1897 |
+
},
|
| 1898 |
+
{
|
| 1899 |
+
"epoch": 2.068651169313839,
|
| 1900 |
+
"eval_cosine_accuracy": 0.9255516529083252,
|
| 1901 |
+
"eval_loss": 9.775038719177246,
|
| 1902 |
+
"eval_runtime": 49.8912,
|
| 1903 |
+
"eval_samples_per_second": 379.145,
|
| 1904 |
+
"eval_steps_per_second": 1.483,
|
| 1905 |
+
"step": 59000
|
| 1906 |
+
},
|
| 1907 |
+
{
|
| 1908 |
+
"epoch": 2.086182111426668,
|
| 1909 |
+
"grad_norm": 4.849012851715088,
|
| 1910 |
+
"learning_rate": 1.4207576000762318e-05,
|
| 1911 |
+
"loss": 9.3162,
|
| 1912 |
+
"step": 59500
|
| 1913 |
+
},
|
| 1914 |
+
{
|
| 1915 |
+
"epoch": 2.086182111426668,
|
| 1916 |
+
"eval_cosine_accuracy": 0.9252797961235046,
|
| 1917 |
+
"eval_loss": 9.710329055786133,
|
| 1918 |
+
"eval_runtime": 51.1116,
|
| 1919 |
+
"eval_samples_per_second": 370.092,
|
| 1920 |
+
"eval_steps_per_second": 1.448,
|
| 1921 |
+
"step": 59500
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 2.1037130535394972,
|
| 1925 |
+
"grad_norm": 5.290930271148682,
|
| 1926 |
+
"learning_rate": 1.4040672055917638e-05,
|
| 1927 |
+
"loss": 9.3126,
|
| 1928 |
+
"step": 60000
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 2.1037130535394972,
|
| 1932 |
+
"eval_cosine_accuracy": 0.9270923137664795,
|
| 1933 |
+
"eval_loss": 9.754308700561523,
|
| 1934 |
+
"eval_runtime": 50.0271,
|
| 1935 |
+
"eval_samples_per_second": 378.115,
|
| 1936 |
+
"eval_steps_per_second": 1.479,
|
| 1937 |
+
"step": 60000
|
| 1938 |
+
},
|
| 1939 |
+
{
|
| 1940 |
+
"epoch": 2.1212439956523266,
|
| 1941 |
+
"grad_norm": 4.115792274475098,
|
| 1942 |
+
"learning_rate": 1.3872073278523122e-05,
|
| 1943 |
+
"loss": 9.3147,
|
| 1944 |
+
"step": 60500
|
| 1945 |
+
},
|
| 1946 |
+
{
|
| 1947 |
+
"epoch": 2.1212439956523266,
|
| 1948 |
+
"eval_cosine_accuracy": 0.9268203973770142,
|
| 1949 |
+
"eval_loss": 9.762267112731934,
|
| 1950 |
+
"eval_runtime": 50.7924,
|
| 1951 |
+
"eval_samples_per_second": 372.418,
|
| 1952 |
+
"eval_steps_per_second": 1.457,
|
| 1953 |
+
"step": 60500
|
| 1954 |
+
},
|
| 1955 |
+
{
|
| 1956 |
+
"epoch": 2.1387749377651555,
|
| 1957 |
+
"grad_norm": 7.710477828979492,
|
| 1958 |
+
"learning_rate": 1.3702169565867822e-05,
|
| 1959 |
+
"loss": 9.3054,
|
| 1960 |
+
"step": 61000
|
| 1961 |
+
},
|
| 1962 |
+
{
|
| 1963 |
+
"epoch": 2.1387749377651555,
|
| 1964 |
+
"eval_cosine_accuracy": 0.9289954304695129,
|
| 1965 |
+
"eval_loss": 9.76127815246582,
|
| 1966 |
+
"eval_runtime": 49.8928,
|
| 1967 |
+
"eval_samples_per_second": 379.133,
|
| 1968 |
+
"eval_steps_per_second": 1.483,
|
| 1969 |
+
"step": 61000
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 2.156305879877985,
|
| 1973 |
+
"grad_norm": 4.304190158843994,
|
| 1974 |
+
"learning_rate": 1.3531018177546172e-05,
|
| 1975 |
+
"loss": 9.3141,
|
| 1976 |
+
"step": 61500
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 2.156305879877985,
|
| 1980 |
+
"eval_cosine_accuracy": 0.9246907234191895,
|
| 1981 |
+
"eval_loss": 9.774465560913086,
|
| 1982 |
+
"eval_runtime": 51.0637,
|
| 1983 |
+
"eval_samples_per_second": 370.439,
|
| 1984 |
+
"eval_steps_per_second": 1.449,
|
| 1985 |
+
"step": 61500
|
| 1986 |
+
},
|
| 1987 |
+
{
|
| 1988 |
+
"epoch": 2.1738368219908137,
|
| 1989 |
+
"grad_norm": 4.020765781402588,
|
| 1990 |
+
"learning_rate": 1.3359022625477911e-05,
|
| 1991 |
+
"loss": 9.3112,
|
| 1992 |
+
"step": 62000
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 2.1738368219908137,
|
| 1996 |
+
"eval_cosine_accuracy": 0.9270016551017761,
|
| 1997 |
+
"eval_loss": 9.729013442993164,
|
| 1998 |
+
"eval_runtime": 50.8117,
|
| 1999 |
+
"eval_samples_per_second": 372.277,
|
| 2000 |
+
"eval_steps_per_second": 1.456,
|
| 2001 |
+
"step": 62000
|
| 2002 |
+
},
|
| 2003 |
+
{
|
| 2004 |
+
"epoch": 2.191367764103643,
|
| 2005 |
+
"grad_norm": 4.387321949005127,
|
| 2006 |
+
"learning_rate": 1.3185551532702006e-05,
|
| 2007 |
+
"loss": 9.3057,
|
| 2008 |
+
"step": 62500
|
| 2009 |
+
},
|
| 2010 |
+
{
|
| 2011 |
+
"epoch": 2.191367764103643,
|
| 2012 |
+
"eval_cosine_accuracy": 0.9264126420021057,
|
| 2013 |
+
"eval_loss": 9.764975547790527,
|
| 2014 |
+
"eval_runtime": 50.1453,
|
| 2015 |
+
"eval_samples_per_second": 377.224,
|
| 2016 |
+
"eval_steps_per_second": 1.476,
|
| 2017 |
+
"step": 62500
|
| 2018 |
+
},
|
| 2019 |
+
{
|
| 2020 |
+
"epoch": 2.208898706216472,
|
| 2021 |
+
"grad_norm": 6.738999843597412,
|
| 2022 |
+
"learning_rate": 1.3011006870749832e-05,
|
| 2023 |
+
"loss": 9.3017,
|
| 2024 |
+
"step": 63000
|
| 2025 |
+
},
|
| 2026 |
+
{
|
| 2027 |
+
"epoch": 2.208898706216472,
|
| 2028 |
+
"eval_cosine_accuracy": 0.9244641661643982,
|
| 2029 |
+
"eval_loss": 9.818182945251465,
|
| 2030 |
+
"eval_runtime": 50.7735,
|
| 2031 |
+
"eval_samples_per_second": 372.556,
|
| 2032 |
+
"eval_steps_per_second": 1.457,
|
| 2033 |
+
"step": 63000
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 2.2264296483293013,
|
| 2037 |
+
"grad_norm": 4.697142601013184,
|
| 2038 |
+
"learning_rate": 1.2835447463271553e-05,
|
| 2039 |
+
"loss": 9.3121,
|
| 2040 |
+
"step": 63500
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 2.2264296483293013,
|
| 2044 |
+
"eval_cosine_accuracy": 0.925143837928772,
|
| 2045 |
+
"eval_loss": 9.72312068939209,
|
| 2046 |
+
"eval_runtime": 49.9392,
|
| 2047 |
+
"eval_samples_per_second": 378.78,
|
| 2048 |
+
"eval_steps_per_second": 1.482,
|
| 2049 |
+
"step": 63500
|
| 2050 |
+
},
|
| 2051 |
+
{
|
| 2052 |
+
"epoch": 2.2439605904421303,
|
| 2053 |
+
"grad_norm": 4.478457927703857,
|
| 2054 |
+
"learning_rate": 1.265928642005434e-05,
|
| 2055 |
+
"loss": 9.3064,
|
| 2056 |
+
"step": 64000
|
| 2057 |
+
},
|
| 2058 |
+
{
|
| 2059 |
+
"epoch": 2.2439605904421303,
|
| 2060 |
+
"eval_cosine_accuracy": 0.9246907234191895,
|
| 2061 |
+
"eval_loss": 9.751876831054688,
|
| 2062 |
+
"eval_runtime": 50.7296,
|
| 2063 |
+
"eval_samples_per_second": 372.879,
|
| 2064 |
+
"eval_steps_per_second": 1.459,
|
| 2065 |
+
"step": 64000
|
| 2066 |
+
},
|
| 2067 |
+
{
|
| 2068 |
+
"epoch": 2.2614915325549596,
|
| 2069 |
+
"grad_norm": 4.6226420402526855,
|
| 2070 |
+
"learning_rate": 1.2481877073072293e-05,
|
| 2071 |
+
"loss": 9.3,
|
| 2072 |
+
"step": 64500
|
| 2073 |
+
},
|
| 2074 |
+
{
|
| 2075 |
+
"epoch": 2.2614915325549596,
|
| 2076 |
+
"eval_cosine_accuracy": 0.927545428276062,
|
| 2077 |
+
"eval_loss": 9.81689739227295,
|
| 2078 |
+
"eval_runtime": 50.0777,
|
| 2079 |
+
"eval_samples_per_second": 377.733,
|
| 2080 |
+
"eval_steps_per_second": 1.478,
|
| 2081 |
+
"step": 64500
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 2.2790224746677885,
|
| 2085 |
+
"grad_norm": 3.7564337253570557,
|
| 2086 |
+
"learning_rate": 1.2303631303669323e-05,
|
| 2087 |
+
"loss": 9.3025,
|
| 2088 |
+
"step": 65000
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 2.2790224746677885,
|
| 2092 |
+
"eval_cosine_accuracy": 0.9258235692977905,
|
| 2093 |
+
"eval_loss": 9.861968994140625,
|
| 2094 |
+
"eval_runtime": 50.4566,
|
| 2095 |
+
"eval_samples_per_second": 374.896,
|
| 2096 |
+
"eval_steps_per_second": 1.467,
|
| 2097 |
+
"step": 65000
|
| 2098 |
+
},
|
| 2099 |
+
{
|
| 2100 |
+
"epoch": 2.296553416780618,
|
| 2101 |
+
"grad_norm": 5.977648735046387,
|
| 2102 |
+
"learning_rate": 1.2124967961794175e-05,
|
| 2103 |
+
"loss": 9.3092,
|
| 2104 |
+
"step": 65500
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"epoch": 2.296553416780618,
|
| 2108 |
+
"eval_cosine_accuracy": 0.9236485362052917,
|
| 2109 |
+
"eval_loss": 9.838058471679688,
|
| 2110 |
+
"eval_runtime": 49.8177,
|
| 2111 |
+
"eval_samples_per_second": 379.705,
|
| 2112 |
+
"eval_steps_per_second": 1.485,
|
| 2113 |
+
"step": 65500
|
| 2114 |
+
},
|
| 2115 |
+
{
|
| 2116 |
+
"epoch": 2.3140843588934468,
|
| 2117 |
+
"grad_norm": 4.86762809753418,
|
| 2118 |
+
"learning_rate": 1.1945231193754782e-05,
|
| 2119 |
+
"loss": 9.2985,
|
| 2120 |
+
"step": 66000
|
| 2121 |
+
},
|
| 2122 |
+
{
|
| 2123 |
+
"epoch": 2.3140843588934468,
|
| 2124 |
+
"eval_cosine_accuracy": 0.9271829128265381,
|
| 2125 |
+
"eval_loss": 9.8118257522583,
|
| 2126 |
+
"eval_runtime": 50.107,
|
| 2127 |
+
"eval_samples_per_second": 377.512,
|
| 2128 |
+
"eval_steps_per_second": 1.477,
|
| 2129 |
+
"step": 66000
|
| 2130 |
+
},
|
| 2131 |
+
{
|
| 2132 |
+
"epoch": 2.331615301006276,
|
| 2133 |
+
"grad_norm": 4.666446208953857,
|
| 2134 |
+
"learning_rate": 1.1764838859409559e-05,
|
| 2135 |
+
"loss": 9.3004,
|
| 2136 |
+
"step": 66500
|
| 2137 |
+
},
|
| 2138 |
+
{
|
| 2139 |
+
"epoch": 2.331615301006276,
|
| 2140 |
+
"eval_cosine_accuracy": 0.9282703995704651,
|
| 2141 |
+
"eval_loss": 9.785202026367188,
|
| 2142 |
+
"eval_runtime": 50.998,
|
| 2143 |
+
"eval_samples_per_second": 370.916,
|
| 2144 |
+
"eval_steps_per_second": 1.451,
|
| 2145 |
+
"step": 66500
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 2.3491462431191055,
|
| 2149 |
+
"grad_norm": 5.578094959259033,
|
| 2150 |
+
"learning_rate": 1.158385175314459e-05,
|
| 2151 |
+
"loss": 9.2944,
|
| 2152 |
+
"step": 67000
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 2.3491462431191055,
|
| 2156 |
+
"eval_cosine_accuracy": 0.9270923137664795,
|
| 2157 |
+
"eval_loss": 9.874638557434082,
|
| 2158 |
+
"eval_runtime": 50.7245,
|
| 2159 |
+
"eval_samples_per_second": 372.916,
|
| 2160 |
+
"eval_steps_per_second": 1.459,
|
| 2161 |
+
"step": 67000
|
| 2162 |
+
},
|
| 2163 |
+
{
|
| 2164 |
+
"epoch": 2.3666771852319344,
|
| 2165 |
+
"grad_norm": 4.742801189422607,
|
| 2166 |
+
"learning_rate": 1.140233086979126e-05,
|
| 2167 |
+
"loss": 9.2935,
|
| 2168 |
+
"step": 67500
|
| 2169 |
+
},
|
| 2170 |
+
{
|
| 2171 |
+
"epoch": 2.3666771852319344,
|
| 2172 |
+
"eval_cosine_accuracy": 0.9260954260826111,
|
| 2173 |
+
"eval_loss": 9.85159683227539,
|
| 2174 |
+
"eval_runtime": 50.3929,
|
| 2175 |
+
"eval_samples_per_second": 375.37,
|
| 2176 |
+
"eval_steps_per_second": 1.468,
|
| 2177 |
+
"step": 67500
|
| 2178 |
+
},
|
| 2179 |
+
{
|
| 2180 |
+
"epoch": 2.3842081273447633,
|
| 2181 |
+
"grad_norm": 5.354438304901123,
|
| 2182 |
+
"learning_rate": 1.1220337384070244e-05,
|
| 2183 |
+
"loss": 9.2944,
|
| 2184 |
+
"step": 68000
|
| 2185 |
+
},
|
| 2186 |
+
{
|
| 2187 |
+
"epoch": 2.3842081273447633,
|
| 2188 |
+
"eval_cosine_accuracy": 0.9259141683578491,
|
| 2189 |
+
"eval_loss": 9.835433959960938,
|
| 2190 |
+
"eval_runtime": 50.8979,
|
| 2191 |
+
"eval_samples_per_second": 371.646,
|
| 2192 |
+
"eval_steps_per_second": 1.454,
|
| 2193 |
+
"step": 68000
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 2.4017390694575926,
|
| 2197 |
+
"grad_norm": 4.104302406311035,
|
| 2198 |
+
"learning_rate": 1.10379326299749e-05,
|
| 2199 |
+
"loss": 9.2971,
|
| 2200 |
+
"step": 68500
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 2.4017390694575926,
|
| 2204 |
+
"eval_cosine_accuracy": 0.9269563555717468,
|
| 2205 |
+
"eval_loss": 9.812936782836914,
|
| 2206 |
+
"eval_runtime": 50.3294,
|
| 2207 |
+
"eval_samples_per_second": 375.844,
|
| 2208 |
+
"eval_steps_per_second": 1.47,
|
| 2209 |
+
"step": 68500
|
| 2210 |
+
},
|
| 2211 |
+
{
|
| 2212 |
+
"epoch": 2.419270011570422,
|
| 2213 |
+
"grad_norm": 7.426550388336182,
|
| 2214 |
+
"learning_rate": 1.085517808010098e-05,
|
| 2215 |
+
"loss": 9.2882,
|
| 2216 |
+
"step": 69000
|
| 2217 |
+
},
|
| 2218 |
+
{
|
| 2219 |
+
"epoch": 2.419270011570422,
|
| 2220 |
+
"eval_cosine_accuracy": 0.9273641705513,
|
| 2221 |
+
"eval_loss": 9.73219108581543,
|
| 2222 |
+
"eval_runtime": 50.9574,
|
| 2223 |
+
"eval_samples_per_second": 371.212,
|
| 2224 |
+
"eval_steps_per_second": 1.452,
|
| 2225 |
+
"step": 69000
|
| 2226 |
+
},
|
| 2227 |
+
{
|
| 2228 |
+
"epoch": 2.436800953683251,
|
| 2229 |
+
"grad_norm": 5.074432849884033,
|
| 2230 |
+
"learning_rate": 1.067213532492961e-05,
|
| 2231 |
+
"loss": 9.2908,
|
| 2232 |
+
"step": 69500
|
| 2233 |
+
},
|
| 2234 |
+
{
|
| 2235 |
+
"epoch": 2.436800953683251,
|
| 2236 |
+
"eval_cosine_accuracy": 0.9276813864707947,
|
| 2237 |
+
"eval_loss": 9.7865629196167,
|
| 2238 |
+
"eval_runtime": 50.0412,
|
| 2239 |
+
"eval_samples_per_second": 378.008,
|
| 2240 |
+
"eval_steps_per_second": 1.479,
|
| 2241 |
+
"step": 69500
|
| 2242 |
+
},
|
| 2243 |
+
{
|
| 2244 |
+
"epoch": 2.4543318957960802,
|
| 2245 |
+
"grad_norm": 4.95829439163208,
|
| 2246 |
+
"learning_rate": 1.0489232775639058e-05,
|
| 2247 |
+
"loss": 9.2878,
|
| 2248 |
+
"step": 70000
|
| 2249 |
+
},
|
| 2250 |
+
{
|
| 2251 |
+
"epoch": 2.4543318957960802,
|
| 2252 |
+
"eval_cosine_accuracy": 0.9255516529083252,
|
| 2253 |
+
"eval_loss": 9.95303726196289,
|
| 2254 |
+
"eval_runtime": 49.7898,
|
| 2255 |
+
"eval_samples_per_second": 379.917,
|
| 2256 |
+
"eval_steps_per_second": 1.486,
|
| 2257 |
+
"step": 70000
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 2.471862837908909,
|
| 2261 |
+
"grad_norm": 5.678023815155029,
|
| 2262 |
+
"learning_rate": 1.030579901686801e-05,
|
| 2263 |
+
"loss": 9.2848,
|
| 2264 |
+
"step": 70500
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 2.471862837908909,
|
| 2268 |
+
"eval_cosine_accuracy": 0.9258235692977905,
|
| 2269 |
+
"eval_loss": 9.936039924621582,
|
| 2270 |
+
"eval_runtime": 52.6665,
|
| 2271 |
+
"eval_samples_per_second": 359.166,
|
| 2272 |
+
"eval_steps_per_second": 1.405,
|
| 2273 |
+
"step": 70500
|
| 2274 |
+
},
|
| 2275 |
+
{
|
| 2276 |
+
"epoch": 2.4893937800217385,
|
| 2277 |
+
"grad_norm": 4.180984973907471,
|
| 2278 |
+
"learning_rate": 1.0122262200150708e-05,
|
| 2279 |
+
"loss": 9.2884,
|
| 2280 |
+
"step": 71000
|
| 2281 |
+
},
|
| 2282 |
+
{
|
| 2283 |
+
"epoch": 2.4893937800217385,
|
| 2284 |
+
"eval_cosine_accuracy": 0.923875093460083,
|
| 2285 |
+
"eval_loss": 9.980805397033691,
|
| 2286 |
+
"eval_runtime": 52.5007,
|
| 2287 |
+
"eval_samples_per_second": 360.3,
|
| 2288 |
+
"eval_steps_per_second": 1.41,
|
| 2289 |
+
"step": 71000
|
| 2290 |
+
},
|
| 2291 |
+
{
|
| 2292 |
+
"epoch": 2.5069247221345674,
|
| 2293 |
+
"grad_norm": 5.1026997566223145,
|
| 2294 |
+
"learning_rate": 9.938684179601618e-06,
|
| 2295 |
+
"loss": 9.2918,
|
| 2296 |
+
"step": 71500
|
| 2297 |
+
},
|
| 2298 |
+
{
|
| 2299 |
+
"epoch": 2.5069247221345674,
|
| 2300 |
+
"eval_cosine_accuracy": 0.9265032410621643,
|
| 2301 |
+
"eval_loss": 9.81791877746582,
|
| 2302 |
+
"eval_runtime": 51.6011,
|
| 2303 |
+
"eval_samples_per_second": 366.581,
|
| 2304 |
+
"eval_steps_per_second": 1.434,
|
| 2305 |
+
"step": 71500
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 2.5244556642473968,
|
| 2309 |
+
"grad_norm": 4.336660861968994,
|
| 2310 |
+
"learning_rate": 9.755493876191022e-06,
|
| 2311 |
+
"loss": 9.2926,
|
| 2312 |
+
"step": 72000
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 2.5244556642473968,
|
| 2316 |
+
"eval_cosine_accuracy": 0.9245094656944275,
|
| 2317 |
+
"eval_loss": 9.875222206115723,
|
| 2318 |
+
"eval_runtime": 51.7235,
|
| 2319 |
+
"eval_samples_per_second": 365.714,
|
| 2320 |
+
"eval_steps_per_second": 1.431,
|
| 2321 |
+
"step": 72000
|
| 2322 |
+
},
|
| 2323 |
+
{
|
| 2324 |
+
"epoch": 2.5419866063602257,
|
| 2325 |
+
"grad_norm": 5.1100568771362305,
|
| 2326 |
+
"learning_rate": 9.572018818243529e-06,
|
| 2327 |
+
"loss": 9.2852,
|
| 2328 |
+
"step": 72500
|
| 2329 |
+
},
|
| 2330 |
+
{
|
| 2331 |
+
"epoch": 2.5419866063602257,
|
| 2332 |
+
"eval_cosine_accuracy": 0.9236485362052917,
|
| 2333 |
+
"eval_loss": 9.831581115722656,
|
| 2334 |
+
"eval_runtime": 53.7517,
|
| 2335 |
+
"eval_samples_per_second": 351.914,
|
| 2336 |
+
"eval_steps_per_second": 1.377,
|
| 2337 |
+
"step": 72500
|
| 2338 |
+
},
|
| 2339 |
+
{
|
| 2340 |
+
"epoch": 2.559517548473055,
|
| 2341 |
+
"grad_norm": 7.713298320770264,
|
| 2342 |
+
"learning_rate": 9.388687995101228e-06,
|
| 2343 |
+
"loss": 9.2799,
|
| 2344 |
+
"step": 73000
|
| 2345 |
+
},
|
| 2346 |
+
{
|
| 2347 |
+
"epoch": 2.559517548473055,
|
| 2348 |
+
"eval_cosine_accuracy": 0.9274094700813293,
|
| 2349 |
+
"eval_loss": 9.851910591125488,
|
| 2350 |
+
"eval_runtime": 52.3587,
|
| 2351 |
+
"eval_samples_per_second": 361.277,
|
| 2352 |
+
"eval_steps_per_second": 1.413,
|
| 2353 |
+
"step": 73000
|
| 2354 |
+
},
|
| 2355 |
+
{
|
| 2356 |
+
"epoch": 2.577048490585884,
|
| 2357 |
+
"grad_norm": 5.331020355224609,
|
| 2358 |
+
"learning_rate": 9.205563191456278e-06,
|
| 2359 |
+
"loss": 9.2793,
|
| 2360 |
+
"step": 73500
|
| 2361 |
+
},
|
| 2362 |
+
{
|
| 2363 |
+
"epoch": 2.577048490585884,
|
| 2364 |
+
"eval_cosine_accuracy": 0.9262313842773438,
|
| 2365 |
+
"eval_loss": 9.931674003601074,
|
| 2366 |
+
"eval_runtime": 51.1059,
|
| 2367 |
+
"eval_samples_per_second": 370.133,
|
| 2368 |
+
"eval_steps_per_second": 1.448,
|
| 2369 |
+
"step": 73500
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 2.5945794326987133,
|
| 2373 |
+
"grad_norm": 4.575249671936035,
|
| 2374 |
+
"learning_rate": 9.02307152853098e-06,
|
| 2375 |
+
"loss": 9.2836,
|
| 2376 |
+
"step": 74000
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 2.5945794326987133,
|
| 2380 |
+
"eval_cosine_accuracy": 0.926276683807373,
|
| 2381 |
+
"eval_loss": 9.854629516601562,
|
| 2382 |
+
"eval_runtime": 52.212,
|
| 2383 |
+
"eval_samples_per_second": 362.292,
|
| 2384 |
+
"eval_steps_per_second": 1.417,
|
| 2385 |
+
"step": 74000
|
| 2386 |
+
},
|
| 2387 |
+
{
|
| 2388 |
+
"epoch": 2.612110374811542,
|
| 2389 |
+
"grad_norm": 7.449720859527588,
|
| 2390 |
+
"learning_rate": 8.84054309928317e-06,
|
| 2391 |
+
"loss": 9.277,
|
| 2392 |
+
"step": 74500
|
| 2393 |
+
},
|
| 2394 |
+
{
|
| 2395 |
+
"epoch": 2.612110374811542,
|
| 2396 |
+
"eval_cosine_accuracy": 0.9250079393386841,
|
| 2397 |
+
"eval_loss": 9.950189590454102,
|
| 2398 |
+
"eval_runtime": 51.6491,
|
| 2399 |
+
"eval_samples_per_second": 366.24,
|
| 2400 |
+
"eval_steps_per_second": 1.433,
|
| 2401 |
+
"step": 74500
|
| 2402 |
+
},
|
| 2403 |
+
{
|
| 2404 |
+
"epoch": 2.6296413169243715,
|
| 2405 |
+
"grad_norm": 12.740141868591309,
|
| 2406 |
+
"learning_rate": 8.658405420954641e-06,
|
| 2407 |
+
"loss": 9.2766,
|
| 2408 |
+
"step": 75000
|
| 2409 |
+
},
|
| 2410 |
+
{
|
| 2411 |
+
"epoch": 2.6296413169243715,
|
| 2412 |
+
"eval_cosine_accuracy": 0.9243735671043396,
|
| 2413 |
+
"eval_loss": 9.890884399414062,
|
| 2414 |
+
"eval_runtime": 52.665,
|
| 2415 |
+
"eval_samples_per_second": 359.176,
|
| 2416 |
+
"eval_steps_per_second": 1.405,
|
| 2417 |
+
"step": 75000
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 2.647172259037201,
|
| 2421 |
+
"grad_norm": 5.653685569763184,
|
| 2422 |
+
"learning_rate": 8.476719876133394e-06,
|
| 2423 |
+
"loss": 9.288,
|
| 2424 |
+
"step": 75500
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 2.647172259037201,
|
| 2428 |
+
"eval_cosine_accuracy": 0.9252344965934753,
|
| 2429 |
+
"eval_loss": 9.862820625305176,
|
| 2430 |
+
"eval_runtime": 51.8074,
|
| 2431 |
+
"eval_samples_per_second": 365.121,
|
| 2432 |
+
"eval_steps_per_second": 1.428,
|
| 2433 |
+
"step": 75500
|
| 2434 |
+
},
|
| 2435 |
+
{
|
| 2436 |
+
"epoch": 2.66470320115003,
|
| 2437 |
+
"grad_norm": 4.039124965667725,
|
| 2438 |
+
"learning_rate": 8.295909486444578e-06,
|
| 2439 |
+
"loss": 9.2757,
|
| 2440 |
+
"step": 76000
|
| 2441 |
+
},
|
| 2442 |
+
{
|
| 2443 |
+
"epoch": 2.66470320115003,
|
| 2444 |
+
"eval_cosine_accuracy": 0.9252344965934753,
|
| 2445 |
+
"eval_loss": 9.927985191345215,
|
| 2446 |
+
"eval_runtime": 52.3107,
|
| 2447 |
+
"eval_samples_per_second": 361.609,
|
| 2448 |
+
"eval_steps_per_second": 1.415,
|
| 2449 |
+
"step": 76000
|
| 2450 |
+
},
|
| 2451 |
+
{
|
| 2452 |
+
"epoch": 2.6822341432628587,
|
| 2453 |
+
"grad_norm": 7.022425651550293,
|
| 2454 |
+
"learning_rate": 8.115310516617183e-06,
|
| 2455 |
+
"loss": 9.2762,
|
| 2456 |
+
"step": 76500
|
| 2457 |
+
},
|
| 2458 |
+
{
|
| 2459 |
+
"epoch": 2.6822341432628587,
|
| 2460 |
+
"eval_cosine_accuracy": 0.9220626354217529,
|
| 2461 |
+
"eval_loss": 9.985993385314941,
|
| 2462 |
+
"eval_runtime": 51.871,
|
| 2463 |
+
"eval_samples_per_second": 364.674,
|
| 2464 |
+
"eval_steps_per_second": 1.427,
|
| 2465 |
+
"step": 76500
|
| 2466 |
+
},
|
| 2467 |
+
{
|
| 2468 |
+
"epoch": 2.699765085375688,
|
| 2469 |
+
"grad_norm": 4.829256057739258,
|
| 2470 |
+
"learning_rate": 7.935346709811144e-06,
|
| 2471 |
+
"loss": 9.2738,
|
| 2472 |
+
"step": 77000
|
| 2473 |
+
},
|
| 2474 |
+
{
|
| 2475 |
+
"epoch": 2.699765085375688,
|
| 2476 |
+
"eval_cosine_accuracy": 0.9263219833374023,
|
| 2477 |
+
"eval_loss": 9.855485916137695,
|
| 2478 |
+
"eval_runtime": 51.9673,
|
| 2479 |
+
"eval_samples_per_second": 363.998,
|
| 2480 |
+
"eval_steps_per_second": 1.424,
|
| 2481 |
+
"step": 77000
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 2.7172960274885174,
|
| 2485 |
+
"grad_norm": 5.828886985778809,
|
| 2486 |
+
"learning_rate": 7.756078715993616e-06,
|
| 2487 |
+
"loss": 9.2701,
|
| 2488 |
+
"step": 77500
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 2.7172960274885174,
|
| 2492 |
+
"eval_cosine_accuracy": 0.9263672828674316,
|
| 2493 |
+
"eval_loss": 9.875953674316406,
|
| 2494 |
+
"eval_runtime": 51.0767,
|
| 2495 |
+
"eval_samples_per_second": 370.345,
|
| 2496 |
+
"eval_steps_per_second": 1.449,
|
| 2497 |
+
"step": 77500
|
| 2498 |
+
},
|
| 2499 |
+
{
|
| 2500 |
+
"epoch": 2.7348269696013463,
|
| 2501 |
+
"grad_norm": 7.390114784240723,
|
| 2502 |
+
"learning_rate": 7.5779231794178365e-06,
|
| 2503 |
+
"loss": 9.2768,
|
| 2504 |
+
"step": 78000
|
| 2505 |
+
},
|
| 2506 |
+
{
|
| 2507 |
+
"epoch": 2.7348269696013463,
|
| 2508 |
+
"eval_cosine_accuracy": 0.9248719811439514,
|
| 2509 |
+
"eval_loss": 9.922650337219238,
|
| 2510 |
+
"eval_runtime": 52.4308,
|
| 2511 |
+
"eval_samples_per_second": 360.78,
|
| 2512 |
+
"eval_steps_per_second": 1.411,
|
| 2513 |
+
"step": 78000
|
| 2514 |
+
},
|
| 2515 |
+
{
|
| 2516 |
+
"epoch": 2.7523579117141757,
|
| 2517 |
+
"grad_norm": 4.566089630126953,
|
| 2518 |
+
"learning_rate": 7.400226110489754e-06,
|
| 2519 |
+
"loss": 9.2659,
|
| 2520 |
+
"step": 78500
|
| 2521 |
+
},
|
| 2522 |
+
{
|
| 2523 |
+
"epoch": 2.7523579117141757,
|
| 2524 |
+
"eval_cosine_accuracy": 0.922606348991394,
|
| 2525 |
+
"eval_loss": 9.981720924377441,
|
| 2526 |
+
"eval_runtime": 51.4659,
|
| 2527 |
+
"eval_samples_per_second": 367.544,
|
| 2528 |
+
"eval_steps_per_second": 1.438,
|
| 2529 |
+
"step": 78500
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 2.7698888538270046,
|
| 2533 |
+
"grad_norm": 4.279365539550781,
|
| 2534 |
+
"learning_rate": 7.223405196627571e-06,
|
| 2535 |
+
"loss": 9.2761,
|
| 2536 |
+
"step": 79000
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 2.7698888538270046,
|
| 2540 |
+
"eval_cosine_accuracy": 0.9219266772270203,
|
| 2541 |
+
"eval_loss": 9.957494735717773,
|
| 2542 |
+
"eval_runtime": 51.3189,
|
| 2543 |
+
"eval_samples_per_second": 368.597,
|
| 2544 |
+
"eval_steps_per_second": 1.442,
|
| 2545 |
+
"step": 79000
|
| 2546 |
+
},
|
| 2547 |
+
{
|
| 2548 |
+
"epoch": 2.787419795939834,
|
| 2549 |
+
"grad_norm": 6.982881546020508,
|
| 2550 |
+
"learning_rate": 7.047520028605721e-06,
|
| 2551 |
+
"loss": 9.2738,
|
| 2552 |
+
"step": 79500
|
| 2553 |
+
},
|
| 2554 |
+
{
|
| 2555 |
+
"epoch": 2.787419795939834,
|
| 2556 |
+
"eval_cosine_accuracy": 0.9237844944000244,
|
| 2557 |
+
"eval_loss": 9.943265914916992,
|
| 2558 |
+
"eval_runtime": 52.9031,
|
| 2559 |
+
"eval_samples_per_second": 357.559,
|
| 2560 |
+
"eval_steps_per_second": 1.399,
|
| 2561 |
+
"step": 79500
|
| 2562 |
+
},
|
| 2563 |
+
{
|
| 2564 |
+
"epoch": 2.804950738052663,
|
| 2565 |
+
"grad_norm": 4.843472480773926,
|
| 2566 |
+
"learning_rate": 6.872978629869944e-06,
|
| 2567 |
+
"loss": 9.2684,
|
| 2568 |
+
"step": 80000
|
| 2569 |
+
},
|
| 2570 |
+
{
|
| 2571 |
+
"epoch": 2.804950738052663,
|
| 2572 |
+
"eval_cosine_accuracy": 0.9231954216957092,
|
| 2573 |
+
"eval_loss": 9.946613311767578,
|
| 2574 |
+
"eval_runtime": 51.543,
|
| 2575 |
+
"eval_samples_per_second": 366.994,
|
| 2576 |
+
"eval_steps_per_second": 1.436,
|
| 2577 |
+
"step": 80000
|
| 2578 |
+
},
|
| 2579 |
+
{
|
| 2580 |
+
"epoch": 2.822481680165492,
|
| 2581 |
+
"grad_norm": 5.201897144317627,
|
| 2582 |
+
"learning_rate": 6.699140277933047e-06,
|
| 2583 |
+
"loss": 9.2622,
|
| 2584 |
+
"step": 80500
|
| 2585 |
+
},
|
| 2586 |
+
{
|
| 2587 |
+
"epoch": 2.822481680165492,
|
| 2588 |
+
"eval_cosine_accuracy": 0.9243282675743103,
|
| 2589 |
+
"eval_loss": 9.957930564880371,
|
| 2590 |
+
"eval_runtime": 52.4634,
|
| 2591 |
+
"eval_samples_per_second": 360.556,
|
| 2592 |
+
"eval_steps_per_second": 1.411,
|
| 2593 |
+
"step": 80500
|
| 2594 |
+
},
|
| 2595 |
+
{
|
| 2596 |
+
"epoch": 2.840012622278321,
|
| 2597 |
+
"grad_norm": 4.828953266143799,
|
| 2598 |
+
"learning_rate": 6.5264143554189395e-06,
|
| 2599 |
+
"loss": 9.2627,
|
| 2600 |
+
"step": 81000
|
| 2601 |
+
},
|
| 2602 |
+
{
|
| 2603 |
+
"epoch": 2.840012622278321,
|
| 2604 |
+
"eval_cosine_accuracy": 0.9238297939300537,
|
| 2605 |
+
"eval_loss": 9.985998153686523,
|
| 2606 |
+
"eval_runtime": 50.1627,
|
| 2607 |
+
"eval_samples_per_second": 377.093,
|
| 2608 |
+
"eval_steps_per_second": 1.475,
|
| 2609 |
+
"step": 81000
|
| 2610 |
+
},
|
| 2611 |
+
{
|
| 2612 |
+
"epoch": 2.8575435643911504,
|
| 2613 |
+
"grad_norm": 5.016162872314453,
|
| 2614 |
+
"learning_rate": 6.354859073040731e-06,
|
| 2615 |
+
"loss": 9.2692,
|
| 2616 |
+
"step": 81500
|
| 2617 |
+
},
|
| 2618 |
+
{
|
| 2619 |
+
"epoch": 2.8575435643911504,
|
| 2620 |
+
"eval_cosine_accuracy": 0.9244188666343689,
|
| 2621 |
+
"eval_loss": 9.962631225585938,
|
| 2622 |
+
"eval_runtime": 51.4218,
|
| 2623 |
+
"eval_samples_per_second": 367.86,
|
| 2624 |
+
"eval_steps_per_second": 1.439,
|
| 2625 |
+
"step": 81500
|
| 2626 |
+
},
|
| 2627 |
+
{
|
| 2628 |
+
"epoch": 2.8750745065039798,
|
| 2629 |
+
"grad_norm": 4.72601318359375,
|
| 2630 |
+
"learning_rate": 6.184532246991723e-06,
|
| 2631 |
+
"loss": 9.2648,
|
| 2632 |
+
"step": 82000
|
| 2633 |
+
},
|
| 2634 |
+
{
|
| 2635 |
+
"epoch": 2.8750745065039798,
|
| 2636 |
+
"eval_cosine_accuracy": 0.9252797961235046,
|
| 2637 |
+
"eval_loss": 9.916666984558105,
|
| 2638 |
+
"eval_runtime": 51.5194,
|
| 2639 |
+
"eval_samples_per_second": 367.163,
|
| 2640 |
+
"eval_steps_per_second": 1.436,
|
| 2641 |
+
"step": 82000
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 2.8926054486168087,
|
| 2645 |
+
"grad_norm": 5.308619022369385,
|
| 2646 |
+
"learning_rate": 6.0158280401694855e-06,
|
| 2647 |
+
"loss": 9.2646,
|
| 2648 |
+
"step": 82500
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 2.8926054486168087,
|
| 2652 |
+
"eval_cosine_accuracy": 0.922968864440918,
|
| 2653 |
+
"eval_loss": 9.951600074768066,
|
| 2654 |
+
"eval_runtime": 52.2971,
|
| 2655 |
+
"eval_samples_per_second": 361.703,
|
| 2656 |
+
"eval_steps_per_second": 1.415,
|
| 2657 |
+
"step": 82500
|
| 2658 |
+
},
|
| 2659 |
+
{
|
| 2660 |
+
"epoch": 2.9101363907296376,
|
| 2661 |
+
"grad_norm": 5.347225189208984,
|
| 2662 |
+
"learning_rate": 5.84812715778333e-06,
|
| 2663 |
+
"loss": 9.2647,
|
| 2664 |
+
"step": 83000
|
| 2665 |
+
},
|
| 2666 |
+
{
|
| 2667 |
+
"epoch": 2.9101363907296376,
|
| 2668 |
+
"eval_cosine_accuracy": 0.9237844944000244,
|
| 2669 |
+
"eval_loss": 9.982772827148438,
|
| 2670 |
+
"eval_runtime": 52.3219,
|
| 2671 |
+
"eval_samples_per_second": 361.531,
|
| 2672 |
+
"eval_steps_per_second": 1.414,
|
| 2673 |
+
"step": 83000
|
| 2674 |
+
},
|
| 2675 |
+
{
|
| 2676 |
+
"epoch": 2.927667332842467,
|
| 2677 |
+
"grad_norm": 8.081692695617676,
|
| 2678 |
+
"learning_rate": 5.681825506475825e-06,
|
| 2679 |
+
"loss": 9.2662,
|
| 2680 |
+
"step": 83500
|
| 2681 |
+
},
|
| 2682 |
+
{
|
| 2683 |
+
"epoch": 2.927667332842467,
|
| 2684 |
+
"eval_cosine_accuracy": 0.9234219789505005,
|
| 2685 |
+
"eval_loss": 9.960512161254883,
|
| 2686 |
+
"eval_runtime": 51.8028,
|
| 2687 |
+
"eval_samples_per_second": 365.154,
|
| 2688 |
+
"eval_steps_per_second": 1.428,
|
| 2689 |
+
"step": 83500
|
| 2690 |
+
},
|
| 2691 |
+
{
|
| 2692 |
+
"epoch": 2.9451982749552963,
|
| 2693 |
+
"grad_norm": 4.40275764465332,
|
| 2694 |
+
"learning_rate": 5.516979131903413e-06,
|
| 2695 |
+
"loss": 9.2663,
|
| 2696 |
+
"step": 84000
|
| 2697 |
+
},
|
| 2698 |
+
{
|
| 2699 |
+
"epoch": 2.9451982749552963,
|
| 2700 |
+
"eval_cosine_accuracy": 0.9225157499313354,
|
| 2701 |
+
"eval_loss": 9.993735313415527,
|
| 2702 |
+
"eval_runtime": 52.4009,
|
| 2703 |
+
"eval_samples_per_second": 360.986,
|
| 2704 |
+
"eval_steps_per_second": 1.412,
|
| 2705 |
+
"step": 84000
|
| 2706 |
+
},
|
| 2707 |
+
{
|
| 2708 |
+
"epoch": 2.962729217068125,
|
| 2709 |
+
"grad_norm": 5.724074840545654,
|
| 2710 |
+
"learning_rate": 5.353643589276794e-06,
|
| 2711 |
+
"loss": 9.2607,
|
| 2712 |
+
"step": 84500
|
| 2713 |
+
},
|
| 2714 |
+
{
|
| 2715 |
+
"epoch": 2.962729217068125,
|
| 2716 |
+
"eval_cosine_accuracy": 0.9234219789505005,
|
| 2717 |
+
"eval_loss": 9.945968627929688,
|
| 2718 |
+
"eval_runtime": 51.6537,
|
| 2719 |
+
"eval_samples_per_second": 366.208,
|
| 2720 |
+
"eval_steps_per_second": 1.433,
|
| 2721 |
+
"step": 84500
|
| 2722 |
+
},
|
| 2723 |
+
{
|
| 2724 |
+
"epoch": 2.9802601591809545,
|
| 2725 |
+
"grad_norm": 3.7497336864471436,
|
| 2726 |
+
"learning_rate": 5.1918739246381166e-06,
|
| 2727 |
+
"loss": 9.2544,
|
| 2728 |
+
"step": 85000
|
| 2729 |
+
},
|
| 2730 |
+
{
|
| 2731 |
+
"epoch": 2.9802601591809545,
|
| 2732 |
+
"eval_cosine_accuracy": 0.9244188666343689,
|
| 2733 |
+
"eval_loss": 10.012833595275879,
|
| 2734 |
+
"eval_runtime": 50.0755,
|
| 2735 |
+
"eval_samples_per_second": 377.75,
|
| 2736 |
+
"eval_steps_per_second": 1.478,
|
| 2737 |
+
"step": 85000
|
| 2738 |
+
},
|
| 2739 |
+
{
|
| 2740 |
+
"epoch": 2.9977911012937835,
|
| 2741 |
+
"grad_norm": 10.823856353759766,
|
| 2742 |
+
"learning_rate": 5.031724656309826e-06,
|
| 2743 |
+
"loss": 9.2611,
|
| 2744 |
+
"step": 85500
|
| 2745 |
+
},
|
| 2746 |
+
{
|
| 2747 |
+
"epoch": 2.9977911012937835,
|
| 2748 |
+
"eval_cosine_accuracy": 0.9242376089096069,
|
| 2749 |
+
"eval_loss": 9.941513061523438,
|
| 2750 |
+
"eval_runtime": 51.0298,
|
| 2751 |
+
"eval_samples_per_second": 370.686,
|
| 2752 |
+
"eval_steps_per_second": 1.45,
|
| 2753 |
+
"step": 85500
|
| 2754 |
+
}
|
| 2755 |
+
],
|
| 2756 |
+
"logging_steps": 500,
|
| 2757 |
+
"max_steps": 114084,
|
| 2758 |
+
"num_input_tokens_seen": 0,
|
| 2759 |
+
"num_train_epochs": 4,
|
| 2760 |
+
"save_steps": 500,
|
| 2761 |
+
"stateful_callbacks": {
|
| 2762 |
+
"TrainerControl": {
|
| 2763 |
+
"args": {
|
| 2764 |
+
"should_epoch_stop": false,
|
| 2765 |
+
"should_evaluate": false,
|
| 2766 |
+
"should_log": false,
|
| 2767 |
+
"should_save": true,
|
| 2768 |
+
"should_training_stop": false
|
| 2769 |
+
},
|
| 2770 |
+
"attributes": {}
|
| 2771 |
+
}
|
| 2772 |
+
},
|
| 2773 |
+
"total_flos": 0.0,
|
| 2774 |
+
"train_batch_size": 256,
|
| 2775 |
+
"trial_name": null,
|
| 2776 |
+
"trial_params": null
|
| 2777 |
+
}
|
checkpoint-85563/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:def7ea8b751f2a45aa92e4170ebf2fe876cda9b58de53add768a1afa69b9be2b
|
| 3 |
+
size 5624
|
checkpoint-85563/training_metrics.json
ADDED
|
@@ -0,0 +1,1032 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metrics_history": [
|
| 3 |
+
{
|
| 4 |
+
"epoch": 0.017530942112829143,
|
| 5 |
+
"global_step": 500,
|
| 6 |
+
"eval_loss": 11.649206161499023,
|
| 7 |
+
"eval_cosine_accuracy": 0.9315329194068909
|
| 8 |
+
},
|
| 9 |
+
{
|
| 10 |
+
"epoch": 0.035061884225658285,
|
| 11 |
+
"global_step": 1000,
|
| 12 |
+
"eval_loss": 11.218177795410156,
|
| 13 |
+
"eval_cosine_accuracy": 0.9300376176834106
|
| 14 |
+
},
|
| 15 |
+
{
|
| 16 |
+
"epoch": 0.05259282633848743,
|
| 17 |
+
"global_step": 1500,
|
| 18 |
+
"eval_loss": 9.87814998626709,
|
| 19 |
+
"eval_cosine_accuracy": 0.9178032279014587
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"epoch": 0.07012376845131657,
|
| 23 |
+
"global_step": 2000,
|
| 24 |
+
"eval_loss": 9.138021469116211,
|
| 25 |
+
"eval_cosine_accuracy": 0.9117313623428345
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.08765471056414571,
|
| 29 |
+
"global_step": 2500,
|
| 30 |
+
"eval_loss": 9.000126838684082,
|
| 31 |
+
"eval_cosine_accuracy": 0.9272282123565674
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.10518565267697486,
|
| 35 |
+
"global_step": 3000,
|
| 36 |
+
"eval_loss": 8.966527938842773,
|
| 37 |
+
"eval_cosine_accuracy": 0.9307625889778137
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.122716594789804,
|
| 41 |
+
"global_step": 3500,
|
| 42 |
+
"eval_loss": 8.893427848815918,
|
| 43 |
+
"eval_cosine_accuracy": 0.9352032542228699
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 0.14024753690263314,
|
| 47 |
+
"global_step": 4000,
|
| 48 |
+
"eval_loss": 8.888650894165039,
|
| 49 |
+
"eval_cosine_accuracy": 0.9365172982215881
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.1577784790154623,
|
| 53 |
+
"global_step": 4500,
|
| 54 |
+
"eval_loss": 8.871306419372559,
|
| 55 |
+
"eval_cosine_accuracy": 0.9367438554763794
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.17530942112829143,
|
| 59 |
+
"global_step": 5000,
|
| 60 |
+
"eval_loss": 8.830900192260742,
|
| 61 |
+
"eval_cosine_accuracy": 0.940504789352417
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.19284036324112058,
|
| 65 |
+
"global_step": 5500,
|
| 66 |
+
"eval_loss": 8.825165748596191,
|
| 67 |
+
"eval_cosine_accuracy": 0.9402329325675964
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.2103713053539497,
|
| 71 |
+
"global_step": 6000,
|
| 72 |
+
"eval_loss": 8.791216850280762,
|
| 73 |
+
"eval_cosine_accuracy": 0.941592276096344
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.22790224746677887,
|
| 77 |
+
"global_step": 6500,
|
| 78 |
+
"eval_loss": 8.7944917678833,
|
| 79 |
+
"eval_cosine_accuracy": 0.9428157210350037
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.245433189579608,
|
| 83 |
+
"global_step": 7000,
|
| 84 |
+
"eval_loss": 8.775042533874512,
|
| 85 |
+
"eval_cosine_accuracy": 0.9449000954627991
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.26296413169243715,
|
| 89 |
+
"global_step": 7500,
|
| 90 |
+
"eval_loss": 8.7813720703125,
|
| 91 |
+
"eval_cosine_accuracy": 0.9455344676971436
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.2804950738052663,
|
| 95 |
+
"global_step": 8000,
|
| 96 |
+
"eval_loss": 8.798937797546387,
|
| 97 |
+
"eval_cosine_accuracy": 0.9440391659736633
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.29802601591809547,
|
| 101 |
+
"global_step": 8500,
|
| 102 |
+
"eval_loss": 8.76003360748291,
|
| 103 |
+
"eval_cosine_accuracy": 0.947256326675415
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 0.3155569580309246,
|
| 107 |
+
"global_step": 9000,
|
| 108 |
+
"eval_loss": 8.755824089050293,
|
| 109 |
+
"eval_cosine_accuracy": 0.9472110271453857
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.3330879001437537,
|
| 113 |
+
"global_step": 9500,
|
| 114 |
+
"eval_loss": 8.768818855285645,
|
| 115 |
+
"eval_cosine_accuracy": 0.9468938112258911
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.35061884225658285,
|
| 119 |
+
"global_step": 10000,
|
| 120 |
+
"eval_loss": 8.746337890625,
|
| 121 |
+
"eval_cosine_accuracy": 0.9488422870635986
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.36814978436941204,
|
| 125 |
+
"global_step": 10500,
|
| 126 |
+
"eval_loss": 8.786169052124023,
|
| 127 |
+
"eval_cosine_accuracy": 0.9490688443183899
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 0.38568072648224117,
|
| 131 |
+
"global_step": 11000,
|
| 132 |
+
"eval_loss": 8.785058975219727,
|
| 133 |
+
"eval_cosine_accuracy": 0.9479813575744629
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 0.4032116685950703,
|
| 137 |
+
"global_step": 11500,
|
| 138 |
+
"eval_loss": 8.779532432556152,
|
| 139 |
+
"eval_cosine_accuracy": 0.9496579170227051
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.4207426107078994,
|
| 143 |
+
"global_step": 12000,
|
| 144 |
+
"eval_loss": 8.792497634887695,
|
| 145 |
+
"eval_cosine_accuracy": 0.9478000998497009
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.4382735528207286,
|
| 149 |
+
"global_step": 12500,
|
| 150 |
+
"eval_loss": 8.789593696594238,
|
| 151 |
+
"eval_cosine_accuracy": 0.949793815612793
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.45580449493355774,
|
| 155 |
+
"global_step": 13000,
|
| 156 |
+
"eval_loss": 8.827651977539062,
|
| 157 |
+
"eval_cosine_accuracy": 0.9480266571044922
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.47333543704638686,
|
| 161 |
+
"global_step": 13500,
|
| 162 |
+
"eval_loss": 8.85822868347168,
|
| 163 |
+
"eval_cosine_accuracy": 0.9472110271453857
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.490866379159216,
|
| 167 |
+
"global_step": 14000,
|
| 168 |
+
"eval_loss": 8.843923568725586,
|
| 169 |
+
"eval_cosine_accuracy": 0.9473016262054443
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.5083973212720452,
|
| 173 |
+
"global_step": 14500,
|
| 174 |
+
"eval_loss": 8.850715637207031,
|
| 175 |
+
"eval_cosine_accuracy": 0.9467579126358032
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 0.5259282633848743,
|
| 179 |
+
"global_step": 15000,
|
| 180 |
+
"eval_loss": 8.862038612365723,
|
| 181 |
+
"eval_cosine_accuracy": 0.9481172561645508
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.5434592054977034,
|
| 185 |
+
"global_step": 15500,
|
| 186 |
+
"eval_loss": 8.83780574798584,
|
| 187 |
+
"eval_cosine_accuracy": 0.9488875865936279
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 0.5609901476105326,
|
| 191 |
+
"global_step": 16000,
|
| 192 |
+
"eval_loss": 8.849825859069824,
|
| 193 |
+
"eval_cosine_accuracy": 0.9463953971862793
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.5785210897233617,
|
| 197 |
+
"global_step": 16500,
|
| 198 |
+
"eval_loss": 8.855111122131348,
|
| 199 |
+
"eval_cosine_accuracy": 0.9481625556945801
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.5960520318361909,
|
| 203 |
+
"global_step": 17000,
|
| 204 |
+
"eval_loss": 8.909035682678223,
|
| 205 |
+
"eval_cosine_accuracy": 0.9444469809532166
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.6135829739490201,
|
| 209 |
+
"global_step": 17500,
|
| 210 |
+
"eval_loss": 8.95804500579834,
|
| 211 |
+
"eval_cosine_accuracy": 0.9443563222885132
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.6311139160618492,
|
| 215 |
+
"global_step": 18000,
|
| 216 |
+
"eval_loss": 8.915170669555664,
|
| 217 |
+
"eval_cosine_accuracy": 0.9461235404014587
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.6486448581746783,
|
| 221 |
+
"global_step": 18500,
|
| 222 |
+
"eval_loss": 8.9291353225708,
|
| 223 |
+
"eval_cosine_accuracy": 0.9455344676971436
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 0.6661758002875074,
|
| 227 |
+
"global_step": 19000,
|
| 228 |
+
"eval_loss": 8.916707038879395,
|
| 229 |
+
"eval_cosine_accuracy": 0.9464406967163086
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 0.6837067424003366,
|
| 233 |
+
"global_step": 19500,
|
| 234 |
+
"eval_loss": 8.885634422302246,
|
| 235 |
+
"eval_cosine_accuracy": 0.9478000998497009
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 0.7012376845131657,
|
| 239 |
+
"global_step": 20000,
|
| 240 |
+
"eval_loss": 9.033989906311035,
|
| 241 |
+
"eval_cosine_accuracy": 0.9428157210350037
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.7187686266259948,
|
| 245 |
+
"global_step": 20500,
|
| 246 |
+
"eval_loss": 8.914213180541992,
|
| 247 |
+
"eval_cosine_accuracy": 0.9464859962463379
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.7362995687388241,
|
| 251 |
+
"global_step": 21000,
|
| 252 |
+
"eval_loss": 9.02304458618164,
|
| 253 |
+
"eval_cosine_accuracy": 0.9437219500541687
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.7538305108516532,
|
| 257 |
+
"global_step": 21500,
|
| 258 |
+
"eval_loss": 8.924837112426758,
|
| 259 |
+
"eval_cosine_accuracy": 0.9458063244819641
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 0.7713614529644823,
|
| 263 |
+
"global_step": 22000,
|
| 264 |
+
"eval_loss": 8.93367862701416,
|
| 265 |
+
"eval_cosine_accuracy": 0.9445375800132751
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.7888923950773115,
|
| 269 |
+
"global_step": 22500,
|
| 270 |
+
"eval_loss": 8.977583885192871,
|
| 271 |
+
"eval_cosine_accuracy": 0.9451719522476196
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.8064233371901406,
|
| 275 |
+
"global_step": 23000,
|
| 276 |
+
"eval_loss": 9.053213119506836,
|
| 277 |
+
"eval_cosine_accuracy": 0.9423173069953918
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 0.8239542793029697,
|
| 281 |
+
"global_step": 23500,
|
| 282 |
+
"eval_loss": 9.02741527557373,
|
| 283 |
+
"eval_cosine_accuracy": 0.9439032077789307
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.8414852214157988,
|
| 287 |
+
"global_step": 24000,
|
| 288 |
+
"eval_loss": 9.003716468811035,
|
| 289 |
+
"eval_cosine_accuracy": 0.9431782364845276
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.859016163528628,
|
| 293 |
+
"global_step": 24500,
|
| 294 |
+
"eval_loss": 8.999176025390625,
|
| 295 |
+
"eval_cosine_accuracy": 0.9453079104423523
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.8765471056414572,
|
| 299 |
+
"global_step": 25000,
|
| 300 |
+
"eval_loss": 9.014289855957031,
|
| 301 |
+
"eval_cosine_accuracy": 0.9445375800132751
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 0.8940780477542863,
|
| 305 |
+
"global_step": 25500,
|
| 306 |
+
"eval_loss": 9.024247169494629,
|
| 307 |
+
"eval_cosine_accuracy": 0.9419094920158386
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 0.9116089898671155,
|
| 311 |
+
"global_step": 26000,
|
| 312 |
+
"eval_loss": 9.146790504455566,
|
| 313 |
+
"eval_cosine_accuracy": 0.9403688311576843
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.9291399319799446,
|
| 317 |
+
"global_step": 26500,
|
| 318 |
+
"eval_loss": 9.119534492492676,
|
| 319 |
+
"eval_cosine_accuracy": 0.9405500888824463
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 0.9466708740927737,
|
| 323 |
+
"global_step": 27000,
|
| 324 |
+
"eval_loss": 9.084246635437012,
|
| 325 |
+
"eval_cosine_accuracy": 0.9424079060554504
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.9642018162056029,
|
| 329 |
+
"global_step": 27500,
|
| 330 |
+
"eval_loss": 9.046530723571777,
|
| 331 |
+
"eval_cosine_accuracy": 0.9432688355445862
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.981732758318432,
|
| 335 |
+
"global_step": 28000,
|
| 336 |
+
"eval_loss": 9.10672664642334,
|
| 337 |
+
"eval_cosine_accuracy": 0.9421360492706299
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 0.9992637004312612,
|
| 341 |
+
"global_step": 28500,
|
| 342 |
+
"eval_loss": 9.120959281921387,
|
| 343 |
+
"eval_cosine_accuracy": 0.9398251175880432
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 1.0167946425440904,
|
| 347 |
+
"global_step": 29000,
|
| 348 |
+
"eval_loss": 9.088321685791016,
|
| 349 |
+
"eval_cosine_accuracy": 0.941592276096344
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 1.0343255846569195,
|
| 353 |
+
"global_step": 29500,
|
| 354 |
+
"eval_loss": 9.25493049621582,
|
| 355 |
+
"eval_cosine_accuracy": 0.9386922717094421
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 1.0518565267697486,
|
| 359 |
+
"global_step": 30000,
|
| 360 |
+
"eval_loss": 9.212674140930176,
|
| 361 |
+
"eval_cosine_accuracy": 0.9394626021385193
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 1.0693874688825777,
|
| 365 |
+
"global_step": 30500,
|
| 366 |
+
"eval_loss": 9.155335426330566,
|
| 367 |
+
"eval_cosine_accuracy": 0.9390094876289368
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.0869184109954069,
|
| 371 |
+
"global_step": 31000,
|
| 372 |
+
"eval_loss": 9.130995750427246,
|
| 373 |
+
"eval_cosine_accuracy": 0.9399157166481018
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 1.104449353108236,
|
| 377 |
+
"global_step": 31500,
|
| 378 |
+
"eval_loss": 9.160995483398438,
|
| 379 |
+
"eval_cosine_accuracy": 0.9375141859054565
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 1.1219802952210651,
|
| 383 |
+
"global_step": 32000,
|
| 384 |
+
"eval_loss": 9.125110626220703,
|
| 385 |
+
"eval_cosine_accuracy": 0.9416828751564026
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 1.1395112373338943,
|
| 389 |
+
"global_step": 32500,
|
| 390 |
+
"eval_loss": 9.20176887512207,
|
| 391 |
+
"eval_cosine_accuracy": 0.9401875734329224
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 1.1570421794467234,
|
| 395 |
+
"global_step": 33000,
|
| 396 |
+
"eval_loss": 9.209524154663086,
|
| 397 |
+
"eval_cosine_accuracy": 0.9374235272407532
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 1.1745731215595527,
|
| 401 |
+
"global_step": 33500,
|
| 402 |
+
"eval_loss": 9.227911949157715,
|
| 403 |
+
"eval_cosine_accuracy": 0.9377407431602478
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 1.1921040636723816,
|
| 407 |
+
"global_step": 34000,
|
| 408 |
+
"eval_loss": 9.227869033813477,
|
| 409 |
+
"eval_cosine_accuracy": 0.9348860383033752
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.209635005785211,
|
| 413 |
+
"global_step": 34500,
|
| 414 |
+
"eval_loss": 9.30997085571289,
|
| 415 |
+
"eval_cosine_accuracy": 0.9332094788551331
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.2271659478980401,
|
| 419 |
+
"global_step": 35000,
|
| 420 |
+
"eval_loss": 9.19826889038086,
|
| 421 |
+
"eval_cosine_accuracy": 0.9390094876289368
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 1.2446968900108693,
|
| 425 |
+
"global_step": 35500,
|
| 426 |
+
"eval_loss": 9.263128280639648,
|
| 427 |
+
"eval_cosine_accuracy": 0.9352485537528992
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 1.2622278321236984,
|
| 431 |
+
"global_step": 36000,
|
| 432 |
+
"eval_loss": 9.285823822021484,
|
| 433 |
+
"eval_cosine_accuracy": 0.9351125955581665
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 1.2797587742365275,
|
| 437 |
+
"global_step": 36500,
|
| 438 |
+
"eval_loss": 9.222258567810059,
|
| 439 |
+
"eval_cosine_accuracy": 0.9381938576698303
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 1.2972897163493566,
|
| 443 |
+
"global_step": 37000,
|
| 444 |
+
"eval_loss": 9.2606840133667,
|
| 445 |
+
"eval_cosine_accuracy": 0.9351125955581665
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 1.3148206584621858,
|
| 449 |
+
"global_step": 37500,
|
| 450 |
+
"eval_loss": 9.276906967163086,
|
| 451 |
+
"eval_cosine_accuracy": 0.934478223323822
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.332351600575015,
|
| 455 |
+
"global_step": 38000,
|
| 456 |
+
"eval_loss": 9.382668495178223,
|
| 457 |
+
"eval_cosine_accuracy": 0.9330282211303711
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.349882542687844,
|
| 461 |
+
"global_step": 38500,
|
| 462 |
+
"eval_loss": 9.369087219238281,
|
| 463 |
+
"eval_cosine_accuracy": 0.9343876242637634
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 1.3674134848006732,
|
| 467 |
+
"global_step": 39000,
|
| 468 |
+
"eval_loss": 9.409356117248535,
|
| 469 |
+
"eval_cosine_accuracy": 0.9313516616821289
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 1.3849444269135023,
|
| 473 |
+
"global_step": 39500,
|
| 474 |
+
"eval_loss": 9.410999298095703,
|
| 475 |
+
"eval_cosine_accuracy": 0.9301282167434692
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 1.4024753690263314,
|
| 479 |
+
"global_step": 40000,
|
| 480 |
+
"eval_loss": 9.38805866241455,
|
| 481 |
+
"eval_cosine_accuracy": 0.9344329237937927
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 1.4200063111391605,
|
| 485 |
+
"global_step": 40500,
|
| 486 |
+
"eval_loss": 9.490105628967285,
|
| 487 |
+
"eval_cosine_accuracy": 0.9307172894477844
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 1.4375372532519899,
|
| 491 |
+
"global_step": 41000,
|
| 492 |
+
"eval_loss": 9.269803047180176,
|
| 493 |
+
"eval_cosine_accuracy": 0.9347047805786133
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 1.4550681953648188,
|
| 497 |
+
"global_step": 41500,
|
| 498 |
+
"eval_loss": 9.31067180633545,
|
| 499 |
+
"eval_cosine_accuracy": 0.9356563687324524
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.4725991374776481,
|
| 503 |
+
"global_step": 42000,
|
| 504 |
+
"eval_loss": 9.364194869995117,
|
| 505 |
+
"eval_cosine_accuracy": 0.9329829216003418
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 1.4901300795904773,
|
| 509 |
+
"global_step": 42500,
|
| 510 |
+
"eval_loss": 9.337225914001465,
|
| 511 |
+
"eval_cosine_accuracy": 0.9347047805786133
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 1.5076610217033064,
|
| 515 |
+
"global_step": 43000,
|
| 516 |
+
"eval_loss": 9.429045677185059,
|
| 517 |
+
"eval_cosine_accuracy": 0.9298110604286194
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 1.5251919638161355,
|
| 521 |
+
"global_step": 43500,
|
| 522 |
+
"eval_loss": 9.397970199584961,
|
| 523 |
+
"eval_cosine_accuracy": 0.9316688776016235
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 1.5427229059289647,
|
| 527 |
+
"global_step": 44000,
|
| 528 |
+
"eval_loss": 9.482829093933105,
|
| 529 |
+
"eval_cosine_accuracy": 0.9302188754081726
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 1.5602538480417938,
|
| 533 |
+
"global_step": 44500,
|
| 534 |
+
"eval_loss": 9.449180603027344,
|
| 535 |
+
"eval_cosine_accuracy": 0.928632915019989
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 1.577784790154623,
|
| 539 |
+
"global_step": 45000,
|
| 540 |
+
"eval_loss": 9.525978088378906,
|
| 541 |
+
"eval_cosine_accuracy": 0.9261860251426697
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.595315732267452,
|
| 545 |
+
"global_step": 45500,
|
| 546 |
+
"eval_loss": 9.421846389770508,
|
| 547 |
+
"eval_cosine_accuracy": 0.9293125867843628
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 1.6128466743802812,
|
| 551 |
+
"global_step": 46000,
|
| 552 |
+
"eval_loss": 9.40963077545166,
|
| 553 |
+
"eval_cosine_accuracy": 0.9294938445091248
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 1.6303776164931103,
|
| 557 |
+
"global_step": 46500,
|
| 558 |
+
"eval_loss": 9.574146270751953,
|
| 559 |
+
"eval_cosine_accuracy": 0.926276683807373
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 1.6479085586059394,
|
| 563 |
+
"global_step": 47000,
|
| 564 |
+
"eval_loss": 9.482501029968262,
|
| 565 |
+
"eval_cosine_accuracy": 0.9274548292160034
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 1.6654395007187688,
|
| 569 |
+
"global_step": 47500,
|
| 570 |
+
"eval_loss": 9.456860542297363,
|
| 571 |
+
"eval_cosine_accuracy": 0.9308985471725464
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 1.6829704428315977,
|
| 575 |
+
"global_step": 48000,
|
| 576 |
+
"eval_loss": 9.518174171447754,
|
| 577 |
+
"eval_cosine_accuracy": 0.9290407299995422
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 1.700501384944427,
|
| 581 |
+
"global_step": 48500,
|
| 582 |
+
"eval_loss": 9.55920696258545,
|
| 583 |
+
"eval_cosine_accuracy": 0.9265032410621643
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.718032327057256,
|
| 587 |
+
"global_step": 49000,
|
| 588 |
+
"eval_loss": 9.498641967773438,
|
| 589 |
+
"eval_cosine_accuracy": 0.9295845031738281
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 1.7355632691700853,
|
| 593 |
+
"global_step": 49500,
|
| 594 |
+
"eval_loss": 9.542733192443848,
|
| 595 |
+
"eval_cosine_accuracy": 0.9289047718048096
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 1.7530942112829142,
|
| 599 |
+
"global_step": 50000,
|
| 600 |
+
"eval_loss": 9.607583999633789,
|
| 601 |
+
"eval_cosine_accuracy": 0.9276360273361206
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 1.7706251533957436,
|
| 605 |
+
"global_step": 50500,
|
| 606 |
+
"eval_loss": 9.642268180847168,
|
| 607 |
+
"eval_cosine_accuracy": 0.9270923137664795
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 1.7881560955085727,
|
| 611 |
+
"global_step": 51000,
|
| 612 |
+
"eval_loss": 9.493197441101074,
|
| 613 |
+
"eval_cosine_accuracy": 0.928814172744751
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 1.8056870376214018,
|
| 617 |
+
"global_step": 51500,
|
| 618 |
+
"eval_loss": 9.589746475219727,
|
| 619 |
+
"eval_cosine_accuracy": 0.9274094700813293
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 1.823217979734231,
|
| 623 |
+
"global_step": 52000,
|
| 624 |
+
"eval_loss": 9.655684471130371,
|
| 625 |
+
"eval_cosine_accuracy": 0.9257329106330872
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.84074892184706,
|
| 629 |
+
"global_step": 52500,
|
| 630 |
+
"eval_loss": 9.585236549377441,
|
| 631 |
+
"eval_cosine_accuracy": 0.9289501309394836
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 1.8582798639598892,
|
| 635 |
+
"global_step": 53000,
|
| 636 |
+
"eval_loss": 9.569564819335938,
|
| 637 |
+
"eval_cosine_accuracy": 0.9288594722747803
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 1.8758108060727183,
|
| 641 |
+
"global_step": 53500,
|
| 642 |
+
"eval_loss": 9.577661514282227,
|
| 643 |
+
"eval_cosine_accuracy": 0.9304001331329346
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 1.8933417481855475,
|
| 647 |
+
"global_step": 54000,
|
| 648 |
+
"eval_loss": 9.534945487976074,
|
| 649 |
+
"eval_cosine_accuracy": 0.9263219833374023
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 1.9108726902983766,
|
| 653 |
+
"global_step": 54500,
|
| 654 |
+
"eval_loss": 9.485892295837402,
|
| 655 |
+
"eval_cosine_accuracy": 0.9293125867843628
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 1.928403632411206,
|
| 659 |
+
"global_step": 55000,
|
| 660 |
+
"eval_loss": 9.556325912475586,
|
| 661 |
+
"eval_cosine_accuracy": 0.9282703995704651
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.9459345745240348,
|
| 665 |
+
"global_step": 55500,
|
| 666 |
+
"eval_loss": 9.641026496887207,
|
| 667 |
+
"eval_cosine_accuracy": 0.9248266816139221
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.9634655166368642,
|
| 671 |
+
"global_step": 56000,
|
| 672 |
+
"eval_loss": 9.674759864807129,
|
| 673 |
+
"eval_cosine_accuracy": 0.924101710319519
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 1.980996458749693,
|
| 677 |
+
"global_step": 56500,
|
| 678 |
+
"eval_loss": 9.623557090759277,
|
| 679 |
+
"eval_cosine_accuracy": 0.9255063533782959
|
| 680 |
+
},
|
| 681 |
+
{
|
| 682 |
+
"epoch": 1.9985274008625225,
|
| 683 |
+
"global_step": 57000,
|
| 684 |
+
"eval_loss": 9.694854736328125,
|
| 685 |
+
"eval_cosine_accuracy": 0.9239204525947571
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 2.0160583429753514,
|
| 689 |
+
"global_step": 57500,
|
| 690 |
+
"eval_loss": 9.723149299621582,
|
| 691 |
+
"eval_cosine_accuracy": 0.9253250956535339
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 2.0335892850881807,
|
| 695 |
+
"global_step": 58000,
|
| 696 |
+
"eval_loss": 9.7728853225708,
|
| 697 |
+
"eval_cosine_accuracy": 0.9257782697677612
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 2.0511202272010096,
|
| 701 |
+
"global_step": 58500,
|
| 702 |
+
"eval_loss": 9.745859146118164,
|
| 703 |
+
"eval_cosine_accuracy": 0.926457941532135
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 2.068651169313839,
|
| 707 |
+
"global_step": 59000,
|
| 708 |
+
"eval_loss": 9.775038719177246,
|
| 709 |
+
"eval_cosine_accuracy": 0.9255516529083252
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 2.086182111426668,
|
| 713 |
+
"global_step": 59500,
|
| 714 |
+
"eval_loss": 9.710329055786133,
|
| 715 |
+
"eval_cosine_accuracy": 0.9252797961235046
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 2.1037130535394972,
|
| 719 |
+
"global_step": 60000,
|
| 720 |
+
"eval_loss": 9.754308700561523,
|
| 721 |
+
"eval_cosine_accuracy": 0.9270923137664795
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 2.1212439956523266,
|
| 725 |
+
"global_step": 60500,
|
| 726 |
+
"eval_loss": 9.762267112731934,
|
| 727 |
+
"eval_cosine_accuracy": 0.9268203973770142
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 2.1387749377651555,
|
| 731 |
+
"global_step": 61000,
|
| 732 |
+
"eval_loss": 9.76127815246582,
|
| 733 |
+
"eval_cosine_accuracy": 0.9289954304695129
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 2.156305879877985,
|
| 737 |
+
"global_step": 61500,
|
| 738 |
+
"eval_loss": 9.774465560913086,
|
| 739 |
+
"eval_cosine_accuracy": 0.9246907234191895
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 2.1738368219908137,
|
| 743 |
+
"global_step": 62000,
|
| 744 |
+
"eval_loss": 9.729013442993164,
|
| 745 |
+
"eval_cosine_accuracy": 0.9270016551017761
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 2.191367764103643,
|
| 749 |
+
"global_step": 62500,
|
| 750 |
+
"eval_loss": 9.764975547790527,
|
| 751 |
+
"eval_cosine_accuracy": 0.9264126420021057
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 2.208898706216472,
|
| 755 |
+
"global_step": 63000,
|
| 756 |
+
"eval_loss": 9.818182945251465,
|
| 757 |
+
"eval_cosine_accuracy": 0.9244641661643982
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 2.2264296483293013,
|
| 761 |
+
"global_step": 63500,
|
| 762 |
+
"eval_loss": 9.72312068939209,
|
| 763 |
+
"eval_cosine_accuracy": 0.925143837928772
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 2.2439605904421303,
|
| 767 |
+
"global_step": 64000,
|
| 768 |
+
"eval_loss": 9.751876831054688,
|
| 769 |
+
"eval_cosine_accuracy": 0.9246907234191895
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 2.2614915325549596,
|
| 773 |
+
"global_step": 64500,
|
| 774 |
+
"eval_loss": 9.81689739227295,
|
| 775 |
+
"eval_cosine_accuracy": 0.927545428276062
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 2.2790224746677885,
|
| 779 |
+
"global_step": 65000,
|
| 780 |
+
"eval_loss": 9.861968994140625,
|
| 781 |
+
"eval_cosine_accuracy": 0.9258235692977905
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 2.296553416780618,
|
| 785 |
+
"global_step": 65500,
|
| 786 |
+
"eval_loss": 9.838058471679688,
|
| 787 |
+
"eval_cosine_accuracy": 0.9236485362052917
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 2.3140843588934468,
|
| 791 |
+
"global_step": 66000,
|
| 792 |
+
"eval_loss": 9.8118257522583,
|
| 793 |
+
"eval_cosine_accuracy": 0.9271829128265381
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 2.331615301006276,
|
| 797 |
+
"global_step": 66500,
|
| 798 |
+
"eval_loss": 9.785202026367188,
|
| 799 |
+
"eval_cosine_accuracy": 0.9282703995704651
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 2.3491462431191055,
|
| 803 |
+
"global_step": 67000,
|
| 804 |
+
"eval_loss": 9.874638557434082,
|
| 805 |
+
"eval_cosine_accuracy": 0.9270923137664795
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 2.3666771852319344,
|
| 809 |
+
"global_step": 67500,
|
| 810 |
+
"eval_loss": 9.85159683227539,
|
| 811 |
+
"eval_cosine_accuracy": 0.9260954260826111
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 2.3842081273447633,
|
| 815 |
+
"global_step": 68000,
|
| 816 |
+
"eval_loss": 9.835433959960938,
|
| 817 |
+
"eval_cosine_accuracy": 0.9259141683578491
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 2.4017390694575926,
|
| 821 |
+
"global_step": 68500,
|
| 822 |
+
"eval_loss": 9.812936782836914,
|
| 823 |
+
"eval_cosine_accuracy": 0.9269563555717468
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 2.419270011570422,
|
| 827 |
+
"global_step": 69000,
|
| 828 |
+
"eval_loss": 9.73219108581543,
|
| 829 |
+
"eval_cosine_accuracy": 0.9273641705513
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 2.436800953683251,
|
| 833 |
+
"global_step": 69500,
|
| 834 |
+
"eval_loss": 9.7865629196167,
|
| 835 |
+
"eval_cosine_accuracy": 0.9276813864707947
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 2.4543318957960802,
|
| 839 |
+
"global_step": 70000,
|
| 840 |
+
"eval_loss": 9.95303726196289,
|
| 841 |
+
"eval_cosine_accuracy": 0.9255516529083252
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 2.471862837908909,
|
| 845 |
+
"global_step": 70500,
|
| 846 |
+
"eval_loss": 9.936039924621582,
|
| 847 |
+
"eval_cosine_accuracy": 0.9258235692977905
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 2.4893937800217385,
|
| 851 |
+
"global_step": 71000,
|
| 852 |
+
"eval_loss": 9.980805397033691,
|
| 853 |
+
"eval_cosine_accuracy": 0.923875093460083
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 2.5069247221345674,
|
| 857 |
+
"global_step": 71500,
|
| 858 |
+
"eval_loss": 9.81791877746582,
|
| 859 |
+
"eval_cosine_accuracy": 0.9265032410621643
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 2.5244556642473968,
|
| 863 |
+
"global_step": 72000,
|
| 864 |
+
"eval_loss": 9.875222206115723,
|
| 865 |
+
"eval_cosine_accuracy": 0.9245094656944275
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 2.5419866063602257,
|
| 869 |
+
"global_step": 72500,
|
| 870 |
+
"eval_loss": 9.831581115722656,
|
| 871 |
+
"eval_cosine_accuracy": 0.9236485362052917
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 2.559517548473055,
|
| 875 |
+
"global_step": 73000,
|
| 876 |
+
"eval_loss": 9.851910591125488,
|
| 877 |
+
"eval_cosine_accuracy": 0.9274094700813293
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 2.577048490585884,
|
| 881 |
+
"global_step": 73500,
|
| 882 |
+
"eval_loss": 9.931674003601074,
|
| 883 |
+
"eval_cosine_accuracy": 0.9262313842773438
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 2.5945794326987133,
|
| 887 |
+
"global_step": 74000,
|
| 888 |
+
"eval_loss": 9.854629516601562,
|
| 889 |
+
"eval_cosine_accuracy": 0.926276683807373
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 2.612110374811542,
|
| 893 |
+
"global_step": 74500,
|
| 894 |
+
"eval_loss": 9.950189590454102,
|
| 895 |
+
"eval_cosine_accuracy": 0.9250079393386841
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 2.6296413169243715,
|
| 899 |
+
"global_step": 75000,
|
| 900 |
+
"eval_loss": 9.890884399414062,
|
| 901 |
+
"eval_cosine_accuracy": 0.9243735671043396
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 2.647172259037201,
|
| 905 |
+
"global_step": 75500,
|
| 906 |
+
"eval_loss": 9.862820625305176,
|
| 907 |
+
"eval_cosine_accuracy": 0.9252344965934753
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 2.66470320115003,
|
| 911 |
+
"global_step": 76000,
|
| 912 |
+
"eval_loss": 9.927985191345215,
|
| 913 |
+
"eval_cosine_accuracy": 0.9252344965934753
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 2.6822341432628587,
|
| 917 |
+
"global_step": 76500,
|
| 918 |
+
"eval_loss": 9.985993385314941,
|
| 919 |
+
"eval_cosine_accuracy": 0.9220626354217529
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 2.699765085375688,
|
| 923 |
+
"global_step": 77000,
|
| 924 |
+
"eval_loss": 9.855485916137695,
|
| 925 |
+
"eval_cosine_accuracy": 0.9263219833374023
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 2.7172960274885174,
|
| 929 |
+
"global_step": 77500,
|
| 930 |
+
"eval_loss": 9.875953674316406,
|
| 931 |
+
"eval_cosine_accuracy": 0.9263672828674316
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 2.7348269696013463,
|
| 935 |
+
"global_step": 78000,
|
| 936 |
+
"eval_loss": 9.922650337219238,
|
| 937 |
+
"eval_cosine_accuracy": 0.9248719811439514
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 2.7523579117141757,
|
| 941 |
+
"global_step": 78500,
|
| 942 |
+
"eval_loss": 9.981720924377441,
|
| 943 |
+
"eval_cosine_accuracy": 0.922606348991394
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 2.7698888538270046,
|
| 947 |
+
"global_step": 79000,
|
| 948 |
+
"eval_loss": 9.957494735717773,
|
| 949 |
+
"eval_cosine_accuracy": 0.9219266772270203
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 2.787419795939834,
|
| 953 |
+
"global_step": 79500,
|
| 954 |
+
"eval_loss": 9.943265914916992,
|
| 955 |
+
"eval_cosine_accuracy": 0.9237844944000244
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 2.804950738052663,
|
| 959 |
+
"global_step": 80000,
|
| 960 |
+
"eval_loss": 9.946613311767578,
|
| 961 |
+
"eval_cosine_accuracy": 0.9231954216957092
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 2.822481680165492,
|
| 965 |
+
"global_step": 80500,
|
| 966 |
+
"eval_loss": 9.957930564880371,
|
| 967 |
+
"eval_cosine_accuracy": 0.9243282675743103
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 2.840012622278321,
|
| 971 |
+
"global_step": 81000,
|
| 972 |
+
"eval_loss": 9.985998153686523,
|
| 973 |
+
"eval_cosine_accuracy": 0.9238297939300537
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 2.8575435643911504,
|
| 977 |
+
"global_step": 81500,
|
| 978 |
+
"eval_loss": 9.962631225585938,
|
| 979 |
+
"eval_cosine_accuracy": 0.9244188666343689
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 2.8750745065039798,
|
| 983 |
+
"global_step": 82000,
|
| 984 |
+
"eval_loss": 9.916666984558105,
|
| 985 |
+
"eval_cosine_accuracy": 0.9252797961235046
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 2.8926054486168087,
|
| 989 |
+
"global_step": 82500,
|
| 990 |
+
"eval_loss": 9.951600074768066,
|
| 991 |
+
"eval_cosine_accuracy": 0.922968864440918
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 2.9101363907296376,
|
| 995 |
+
"global_step": 83000,
|
| 996 |
+
"eval_loss": 9.982772827148438,
|
| 997 |
+
"eval_cosine_accuracy": 0.9237844944000244
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 2.927667332842467,
|
| 1001 |
+
"global_step": 83500,
|
| 1002 |
+
"eval_loss": 9.960512161254883,
|
| 1003 |
+
"eval_cosine_accuracy": 0.9234219789505005
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 2.9451982749552963,
|
| 1007 |
+
"global_step": 84000,
|
| 1008 |
+
"eval_loss": 9.993735313415527,
|
| 1009 |
+
"eval_cosine_accuracy": 0.9225157499313354
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 2.962729217068125,
|
| 1013 |
+
"global_step": 84500,
|
| 1014 |
+
"eval_loss": 9.945968627929688,
|
| 1015 |
+
"eval_cosine_accuracy": 0.9234219789505005
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 2.9802601591809545,
|
| 1019 |
+
"global_step": 85000,
|
| 1020 |
+
"eval_loss": 10.012833595275879,
|
| 1021 |
+
"eval_cosine_accuracy": 0.9244188666343689
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 2.9977911012937835,
|
| 1025 |
+
"global_step": 85500,
|
| 1026 |
+
"eval_loss": 9.941513061523438,
|
| 1027 |
+
"eval_cosine_accuracy": 0.9242376089096069
|
| 1028 |
+
}
|
| 1029 |
+
],
|
| 1030 |
+
"current_epoch": 3,
|
| 1031 |
+
"global_step": 85563
|
| 1032 |
+
}
|
checkpoint-85563/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|