Training checkpoint - Epoch 2, Step 47030
Browse files- checkpoint-47030/1_Pooling/config.json +10 -0
- checkpoint-47030/config.json +25 -0
- checkpoint-47030/config_sentence_transformers.json +14 -0
- checkpoint-47030/model.safetensors +3 -0
- checkpoint-47030/modules.json +20 -0
- checkpoint-47030/optimizer.pt +3 -0
- checkpoint-47030/rng_state.pth +3 -0
- checkpoint-47030/scaler.pt +3 -0
- checkpoint-47030/scheduler.pt +3 -0
- checkpoint-47030/sentence_bert_config.json +4 -0
- checkpoint-47030/special_tokens_map.json +37 -0
- checkpoint-47030/tokenizer.json +0 -0
- checkpoint-47030/tokenizer_config.json +65 -0
- checkpoint-47030/trainer_state.json +1545 -0
- checkpoint-47030/training_args.bin +3 -0
- checkpoint-47030/training_metrics.json +570 -0
- checkpoint-47030/vocab.txt +0 -0
checkpoint-47030/1_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-47030/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-47030/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-47030/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b216963fd7982d613568d7a651c1e8b24bde87804c7fb529cf2fc9155e290aee
|
| 3 |
+
size 90864192
|
checkpoint-47030/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-47030/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0392671bdc36c239727c7a556008386e85cb6b766baf1aa6cb21e53155421e2f
|
| 3 |
+
size 180607738
|
checkpoint-47030/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:630ff0c30a791bdbd717b7cb1abc8649f8dccd251abad4d11aa81e9f2851401c
|
| 3 |
+
size 14244
|
checkpoint-47030/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b78ff5b1d89703f75e921d99ee8f2028da0595420d9fe7d3cd5996782908bd86
|
| 3 |
+
size 988
|
checkpoint-47030/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c895dd06c926bda3bf8d4f91b44c41df838612b244045317df23257ab37568e1
|
| 3 |
+
size 1064
|
checkpoint-47030/sentence_bert_config.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"max_seq_length": 256,
|
| 3 |
+
"do_lower_case": false
|
| 4 |
+
}
|
checkpoint-47030/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-47030/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-47030/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-47030/trainer_state.json
ADDED
|
@@ -0,0 +1,1545 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 2.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 47030,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 4.2526047203912395e-05,
|
| 14 |
+
"grad_norm": NaN,
|
| 15 |
+
"learning_rate": 0.0,
|
| 16 |
+
"loss": 16.5747,
|
| 17 |
+
"step": 1
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.0212630236019562,
|
| 21 |
+
"grad_norm": 72.24292755126953,
|
| 22 |
+
"learning_rate": 2.4719332578020393e-08,
|
| 23 |
+
"loss": 16.0022,
|
| 24 |
+
"step": 500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0212630236019562,
|
| 28 |
+
"eval_cosine_accuracy": 0.9308079481124878,
|
| 29 |
+
"eval_loss": 11.834263801574707,
|
| 30 |
+
"eval_runtime": 49.7307,
|
| 31 |
+
"eval_samples_per_second": 380.208,
|
| 32 |
+
"eval_steps_per_second": 1.488,
|
| 33 |
+
"step": 500
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.0425260472039124,
|
| 37 |
+
"grad_norm": 90.77424621582031,
|
| 38 |
+
"learning_rate": 4.963801461231515e-08,
|
| 39 |
+
"loss": 15.9413,
|
| 40 |
+
"step": 1000
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.0425260472039124,
|
| 44 |
+
"eval_cosine_accuracy": 0.9313969612121582,
|
| 45 |
+
"eval_loss": 11.764849662780762,
|
| 46 |
+
"eval_runtime": 50.6172,
|
| 47 |
+
"eval_samples_per_second": 373.549,
|
| 48 |
+
"eval_steps_per_second": 1.462,
|
| 49 |
+
"step": 1000
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.0637890708058686,
|
| 53 |
+
"grad_norm": 61.71802520751953,
|
| 54 |
+
"learning_rate": 7.45566966466099e-08,
|
| 55 |
+
"loss": 15.8159,
|
| 56 |
+
"step": 1500
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.0637890708058686,
|
| 60 |
+
"eval_cosine_accuracy": 0.9325751066207886,
|
| 61 |
+
"eval_loss": 11.667285919189453,
|
| 62 |
+
"eval_runtime": 50.0024,
|
| 63 |
+
"eval_samples_per_second": 378.142,
|
| 64 |
+
"eval_steps_per_second": 1.48,
|
| 65 |
+
"step": 1500
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.0850520944078248,
|
| 69 |
+
"grad_norm": 70.19003295898438,
|
| 70 |
+
"learning_rate": 9.947537868090464e-08,
|
| 71 |
+
"loss": 15.6699,
|
| 72 |
+
"step": 2000
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.0850520944078248,
|
| 76 |
+
"eval_cosine_accuracy": 0.934478223323822,
|
| 77 |
+
"eval_loss": 11.543947219848633,
|
| 78 |
+
"eval_runtime": 50.7737,
|
| 79 |
+
"eval_samples_per_second": 372.397,
|
| 80 |
+
"eval_steps_per_second": 1.457,
|
| 81 |
+
"step": 2000
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.106315118009781,
|
| 85 |
+
"grad_norm": 88.77647399902344,
|
| 86 |
+
"learning_rate": 1.2434422335113082e-07,
|
| 87 |
+
"loss": 15.3678,
|
| 88 |
+
"step": 2500
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.106315118009781,
|
| 92 |
+
"eval_cosine_accuracy": 0.9360641837120056,
|
| 93 |
+
"eval_loss": 11.422866821289062,
|
| 94 |
+
"eval_runtime": 49.9051,
|
| 95 |
+
"eval_samples_per_second": 378.879,
|
| 96 |
+
"eval_steps_per_second": 1.483,
|
| 97 |
+
"step": 2500
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.1275781416117372,
|
| 101 |
+
"grad_norm": 60.29121398925781,
|
| 102 |
+
"learning_rate": 1.4926290538542556e-07,
|
| 103 |
+
"loss": 15.0132,
|
| 104 |
+
"step": 3000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.1275781416117372,
|
| 108 |
+
"eval_cosine_accuracy": 0.9384657144546509,
|
| 109 |
+
"eval_loss": 11.268472671508789,
|
| 110 |
+
"eval_runtime": 50.5525,
|
| 111 |
+
"eval_samples_per_second": 374.027,
|
| 112 |
+
"eval_steps_per_second": 1.464,
|
| 113 |
+
"step": 3000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.1488411652136934,
|
| 117 |
+
"grad_norm": 58.82876968383789,
|
| 118 |
+
"learning_rate": 1.741815874197203e-07,
|
| 119 |
+
"loss": 14.5215,
|
| 120 |
+
"step": 3500
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.1488411652136934,
|
| 124 |
+
"eval_cosine_accuracy": 0.941592276096344,
|
| 125 |
+
"eval_loss": 11.027884483337402,
|
| 126 |
+
"eval_runtime": 49.4569,
|
| 127 |
+
"eval_samples_per_second": 382.313,
|
| 128 |
+
"eval_steps_per_second": 1.496,
|
| 129 |
+
"step": 3500
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.1701041888156496,
|
| 133 |
+
"grad_norm": 86.84857177734375,
|
| 134 |
+
"learning_rate": 1.9910026945401506e-07,
|
| 135 |
+
"loss": 14.0143,
|
| 136 |
+
"step": 4000
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.1701041888156496,
|
| 140 |
+
"eval_cosine_accuracy": 0.9445375800132751,
|
| 141 |
+
"eval_loss": 10.77338981628418,
|
| 142 |
+
"eval_runtime": 49.47,
|
| 143 |
+
"eval_samples_per_second": 382.212,
|
| 144 |
+
"eval_steps_per_second": 1.496,
|
| 145 |
+
"step": 4000
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.1913672124176058,
|
| 149 |
+
"grad_norm": 50.29563903808594,
|
| 150 |
+
"learning_rate": 2.2396911412424123e-07,
|
| 151 |
+
"loss": 13.3569,
|
| 152 |
+
"step": 4500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.1913672124176058,
|
| 156 |
+
"eval_cosine_accuracy": 0.9471657276153564,
|
| 157 |
+
"eval_loss": 10.502735137939453,
|
| 158 |
+
"eval_runtime": 50.4762,
|
| 159 |
+
"eval_samples_per_second": 374.592,
|
| 160 |
+
"eval_steps_per_second": 1.466,
|
| 161 |
+
"step": 4500
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.212630236019562,
|
| 165 |
+
"grad_norm": 48.2503776550293,
|
| 166 |
+
"learning_rate": 2.48887796158536e-07,
|
| 167 |
+
"loss": 12.5679,
|
| 168 |
+
"step": 5000
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 0.212630236019562,
|
| 172 |
+
"eval_cosine_accuracy": 0.9499297738075256,
|
| 173 |
+
"eval_loss": 10.223851203918457,
|
| 174 |
+
"eval_runtime": 49.7554,
|
| 175 |
+
"eval_samples_per_second": 380.019,
|
| 176 |
+
"eval_steps_per_second": 1.487,
|
| 177 |
+
"step": 5000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.2338932596215182,
|
| 181 |
+
"grad_norm": 85.99883270263672,
|
| 182 |
+
"learning_rate": 2.7380647819283073e-07,
|
| 183 |
+
"loss": 11.7751,
|
| 184 |
+
"step": 5500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.2338932596215182,
|
| 188 |
+
"eval_cosine_accuracy": 0.9533734917640686,
|
| 189 |
+
"eval_loss": 9.879401206970215,
|
| 190 |
+
"eval_runtime": 50.9383,
|
| 191 |
+
"eval_samples_per_second": 371.194,
|
| 192 |
+
"eval_steps_per_second": 1.453,
|
| 193 |
+
"step": 5500
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.2551562832234744,
|
| 197 |
+
"grad_norm": 30.850263595581055,
|
| 198 |
+
"learning_rate": 2.9872516022712553e-07,
|
| 199 |
+
"loss": 10.924,
|
| 200 |
+
"step": 6000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.2551562832234744,
|
| 204 |
+
"eval_cosine_accuracy": 0.9558656811714172,
|
| 205 |
+
"eval_loss": 9.583606719970703,
|
| 206 |
+
"eval_runtime": 50.1057,
|
| 207 |
+
"eval_samples_per_second": 377.363,
|
| 208 |
+
"eval_steps_per_second": 1.477,
|
| 209 |
+
"step": 6000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.2764193068254306,
|
| 213 |
+
"grad_norm": 27.160547256469727,
|
| 214 |
+
"learning_rate": 3.2359400489735165e-07,
|
| 215 |
+
"loss": 10.207,
|
| 216 |
+
"step": 6500
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.2764193068254306,
|
| 220 |
+
"eval_cosine_accuracy": 0.9583125710487366,
|
| 221 |
+
"eval_loss": 9.295426368713379,
|
| 222 |
+
"eval_runtime": 50.8853,
|
| 223 |
+
"eval_samples_per_second": 371.58,
|
| 224 |
+
"eval_steps_per_second": 1.454,
|
| 225 |
+
"step": 6500
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.2976823304273868,
|
| 229 |
+
"grad_norm": 14.218546867370605,
|
| 230 |
+
"learning_rate": 3.4851268693164645e-07,
|
| 231 |
+
"loss": 9.6073,
|
| 232 |
+
"step": 7000
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.2976823304273868,
|
| 236 |
+
"eval_cosine_accuracy": 0.9617562890052795,
|
| 237 |
+
"eval_loss": 9.070219993591309,
|
| 238 |
+
"eval_runtime": 49.4097,
|
| 239 |
+
"eval_samples_per_second": 382.678,
|
| 240 |
+
"eval_steps_per_second": 1.498,
|
| 241 |
+
"step": 7000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.318945354029343,
|
| 245 |
+
"grad_norm": 20.438467025756836,
|
| 246 |
+
"learning_rate": 3.7343136896594114e-07,
|
| 247 |
+
"loss": 9.3031,
|
| 248 |
+
"step": 7500
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.318945354029343,
|
| 252 |
+
"eval_cosine_accuracy": 0.9646109938621521,
|
| 253 |
+
"eval_loss": 8.921951293945312,
|
| 254 |
+
"eval_runtime": 50.9982,
|
| 255 |
+
"eval_samples_per_second": 370.758,
|
| 256 |
+
"eval_steps_per_second": 1.451,
|
| 257 |
+
"step": 7500
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.3402083776312992,
|
| 261 |
+
"grad_norm": 18.741724014282227,
|
| 262 |
+
"learning_rate": 3.983500510002359e-07,
|
| 263 |
+
"loss": 9.1113,
|
| 264 |
+
"step": 8000
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.3402083776312992,
|
| 268 |
+
"eval_cosine_accuracy": 0.9665141105651855,
|
| 269 |
+
"eval_loss": 8.841903686523438,
|
| 270 |
+
"eval_runtime": 49.583,
|
| 271 |
+
"eval_samples_per_second": 381.34,
|
| 272 |
+
"eval_steps_per_second": 1.492,
|
| 273 |
+
"step": 8000
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.3614714012332554,
|
| 277 |
+
"grad_norm": 14.93420696258545,
|
| 278 |
+
"learning_rate": 4.2321889567046206e-07,
|
| 279 |
+
"loss": 8.9949,
|
| 280 |
+
"step": 8500
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.3614714012332554,
|
| 284 |
+
"eval_cosine_accuracy": 0.9681000709533691,
|
| 285 |
+
"eval_loss": 8.776097297668457,
|
| 286 |
+
"eval_runtime": 50.3732,
|
| 287 |
+
"eval_samples_per_second": 375.359,
|
| 288 |
+
"eval_steps_per_second": 1.469,
|
| 289 |
+
"step": 8500
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.3827344248352116,
|
| 293 |
+
"grad_norm": 12.44970417022705,
|
| 294 |
+
"learning_rate": 4.481375777047568e-07,
|
| 295 |
+
"loss": 8.9394,
|
| 296 |
+
"step": 9000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.3827344248352116,
|
| 300 |
+
"eval_cosine_accuracy": 0.9681000709533691,
|
| 301 |
+
"eval_loss": 8.712888717651367,
|
| 302 |
+
"eval_runtime": 49.3318,
|
| 303 |
+
"eval_samples_per_second": 383.282,
|
| 304 |
+
"eval_steps_per_second": 1.5,
|
| 305 |
+
"step": 9000
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.4039974484371678,
|
| 309 |
+
"grad_norm": 11.820176124572754,
|
| 310 |
+
"learning_rate": 4.7305625973905156e-07,
|
| 311 |
+
"loss": 8.8496,
|
| 312 |
+
"step": 9500
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 0.4039974484371678,
|
| 316 |
+
"eval_cosine_accuracy": 0.968235969543457,
|
| 317 |
+
"eval_loss": 8.671957015991211,
|
| 318 |
+
"eval_runtime": 50.445,
|
| 319 |
+
"eval_samples_per_second": 374.824,
|
| 320 |
+
"eval_steps_per_second": 1.467,
|
| 321 |
+
"step": 9500
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.425260472039124,
|
| 325 |
+
"grad_norm": 19.25648307800293,
|
| 326 |
+
"learning_rate": 4.979251044092777e-07,
|
| 327 |
+
"loss": 8.8057,
|
| 328 |
+
"step": 10000
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.425260472039124,
|
| 332 |
+
"eval_cosine_accuracy": 0.9679641127586365,
|
| 333 |
+
"eval_loss": 8.636693000793457,
|
| 334 |
+
"eval_runtime": 49.6391,
|
| 335 |
+
"eval_samples_per_second": 380.909,
|
| 336 |
+
"eval_steps_per_second": 1.491,
|
| 337 |
+
"step": 10000
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 0.4465234956410802,
|
| 341 |
+
"grad_norm": 11.33238697052002,
|
| 342 |
+
"learning_rate": 5.228437864435725e-07,
|
| 343 |
+
"loss": 8.739,
|
| 344 |
+
"step": 10500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.4465234956410802,
|
| 348 |
+
"eval_cosine_accuracy": 0.9684625267982483,
|
| 349 |
+
"eval_loss": 8.61970329284668,
|
| 350 |
+
"eval_runtime": 50.2316,
|
| 351 |
+
"eval_samples_per_second": 376.416,
|
| 352 |
+
"eval_steps_per_second": 1.473,
|
| 353 |
+
"step": 10500
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.4677865192430364,
|
| 357 |
+
"grad_norm": 12.054763793945312,
|
| 358 |
+
"learning_rate": 5.477624684778672e-07,
|
| 359 |
+
"loss": 8.6886,
|
| 360 |
+
"step": 11000
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.4677865192430364,
|
| 364 |
+
"eval_cosine_accuracy": 0.968598484992981,
|
| 365 |
+
"eval_loss": 8.59864616394043,
|
| 366 |
+
"eval_runtime": 49.6249,
|
| 367 |
+
"eval_samples_per_second": 381.018,
|
| 368 |
+
"eval_steps_per_second": 1.491,
|
| 369 |
+
"step": 11000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.4890495428449926,
|
| 373 |
+
"grad_norm": 12.590415000915527,
|
| 374 |
+
"learning_rate": 5.72681150512162e-07,
|
| 375 |
+
"loss": 8.6431,
|
| 376 |
+
"step": 11500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 0.4890495428449926,
|
| 380 |
+
"eval_cosine_accuracy": 0.9678281545639038,
|
| 381 |
+
"eval_loss": 8.581836700439453,
|
| 382 |
+
"eval_runtime": 50.2665,
|
| 383 |
+
"eval_samples_per_second": 376.155,
|
| 384 |
+
"eval_steps_per_second": 1.472,
|
| 385 |
+
"step": 11500
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 0.5103125664469488,
|
| 389 |
+
"grad_norm": 15.370965957641602,
|
| 390 |
+
"learning_rate": 5.975998325464567e-07,
|
| 391 |
+
"loss": 8.6208,
|
| 392 |
+
"step": 12000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.5103125664469488,
|
| 396 |
+
"eval_cosine_accuracy": 0.968598484992981,
|
| 397 |
+
"eval_loss": 8.53200912475586,
|
| 398 |
+
"eval_runtime": 49.502,
|
| 399 |
+
"eval_samples_per_second": 381.964,
|
| 400 |
+
"eval_steps_per_second": 1.495,
|
| 401 |
+
"step": 12000
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.531575590048905,
|
| 405 |
+
"grad_norm": 10.931774139404297,
|
| 406 |
+
"learning_rate": 6.225185145807515e-07,
|
| 407 |
+
"loss": 8.5638,
|
| 408 |
+
"step": 12500
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.531575590048905,
|
| 412 |
+
"eval_cosine_accuracy": 0.9685078859329224,
|
| 413 |
+
"eval_loss": 8.522621154785156,
|
| 414 |
+
"eval_runtime": 50.2178,
|
| 415 |
+
"eval_samples_per_second": 376.52,
|
| 416 |
+
"eval_steps_per_second": 1.474,
|
| 417 |
+
"step": 12500
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.5528386136508612,
|
| 421 |
+
"grad_norm": 13.296648025512695,
|
| 422 |
+
"learning_rate": 6.474371966150463e-07,
|
| 423 |
+
"loss": 8.539,
|
| 424 |
+
"step": 13000
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 0.5528386136508612,
|
| 428 |
+
"eval_cosine_accuracy": 0.9688250422477722,
|
| 429 |
+
"eval_loss": 8.52730941772461,
|
| 430 |
+
"eval_runtime": 49.7781,
|
| 431 |
+
"eval_samples_per_second": 379.845,
|
| 432 |
+
"eval_steps_per_second": 1.487,
|
| 433 |
+
"step": 13000
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.5741016372528174,
|
| 437 |
+
"grad_norm": 13.039021492004395,
|
| 438 |
+
"learning_rate": 6.72355878649341e-07,
|
| 439 |
+
"loss": 8.4952,
|
| 440 |
+
"step": 13500
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.5741016372528174,
|
| 444 |
+
"eval_cosine_accuracy": 0.9687797427177429,
|
| 445 |
+
"eval_loss": 8.488916397094727,
|
| 446 |
+
"eval_runtime": 50.244,
|
| 447 |
+
"eval_samples_per_second": 376.324,
|
| 448 |
+
"eval_steps_per_second": 1.473,
|
| 449 |
+
"step": 13500
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.5953646608547736,
|
| 453 |
+
"grad_norm": 14.663399696350098,
|
| 454 |
+
"learning_rate": 6.972745606836358e-07,
|
| 455 |
+
"loss": 8.437,
|
| 456 |
+
"step": 14000
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 0.5953646608547736,
|
| 460 |
+
"eval_cosine_accuracy": 0.9692781567573547,
|
| 461 |
+
"eval_loss": 8.479202270507812,
|
| 462 |
+
"eval_runtime": 49.8834,
|
| 463 |
+
"eval_samples_per_second": 379.044,
|
| 464 |
+
"eval_steps_per_second": 1.483,
|
| 465 |
+
"step": 14000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.6166276844567298,
|
| 469 |
+
"grad_norm": 14.649970054626465,
|
| 470 |
+
"learning_rate": 7.221932427179305e-07,
|
| 471 |
+
"loss": 8.4223,
|
| 472 |
+
"step": 14500
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.6166276844567298,
|
| 476 |
+
"eval_cosine_accuracy": 0.9690062999725342,
|
| 477 |
+
"eval_loss": 8.463949203491211,
|
| 478 |
+
"eval_runtime": 50.2893,
|
| 479 |
+
"eval_samples_per_second": 375.985,
|
| 480 |
+
"eval_steps_per_second": 1.471,
|
| 481 |
+
"step": 14500
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 0.637890708058686,
|
| 485 |
+
"grad_norm": 16.208450317382812,
|
| 486 |
+
"learning_rate": 7.471119247522253e-07,
|
| 487 |
+
"loss": 8.367,
|
| 488 |
+
"step": 15000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.637890708058686,
|
| 492 |
+
"eval_cosine_accuracy": 0.9686437845230103,
|
| 493 |
+
"eval_loss": 8.437935829162598,
|
| 494 |
+
"eval_runtime": 49.4396,
|
| 495 |
+
"eval_samples_per_second": 382.446,
|
| 496 |
+
"eval_steps_per_second": 1.497,
|
| 497 |
+
"step": 15000
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 0.6591537316606422,
|
| 501 |
+
"grad_norm": 17.769954681396484,
|
| 502 |
+
"learning_rate": 7.7203060678652e-07,
|
| 503 |
+
"loss": 8.3536,
|
| 504 |
+
"step": 15500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.6591537316606422,
|
| 508 |
+
"eval_cosine_accuracy": 0.9689610004425049,
|
| 509 |
+
"eval_loss": 8.473112106323242,
|
| 510 |
+
"eval_runtime": 50.2825,
|
| 511 |
+
"eval_samples_per_second": 376.035,
|
| 512 |
+
"eval_steps_per_second": 1.472,
|
| 513 |
+
"step": 15500
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.6804167552625984,
|
| 517 |
+
"grad_norm": 17.118555068969727,
|
| 518 |
+
"learning_rate": 7.968994514567462e-07,
|
| 519 |
+
"loss": 8.3289,
|
| 520 |
+
"step": 16000
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.6804167552625984,
|
| 524 |
+
"eval_cosine_accuracy": 0.9688703417778015,
|
| 525 |
+
"eval_loss": 8.445318222045898,
|
| 526 |
+
"eval_runtime": 49.5183,
|
| 527 |
+
"eval_samples_per_second": 381.838,
|
| 528 |
+
"eval_steps_per_second": 1.494,
|
| 529 |
+
"step": 16000
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.7016797788645546,
|
| 533 |
+
"grad_norm": 14.455836296081543,
|
| 534 |
+
"learning_rate": 8.218181334910409e-07,
|
| 535 |
+
"loss": 8.3108,
|
| 536 |
+
"step": 16500
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.7016797788645546,
|
| 540 |
+
"eval_cosine_accuracy": 0.9694594144821167,
|
| 541 |
+
"eval_loss": 8.421878814697266,
|
| 542 |
+
"eval_runtime": 50.3275,
|
| 543 |
+
"eval_samples_per_second": 375.699,
|
| 544 |
+
"eval_steps_per_second": 1.47,
|
| 545 |
+
"step": 16500
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 0.7229428024665108,
|
| 549 |
+
"grad_norm": 13.595857620239258,
|
| 550 |
+
"learning_rate": 8.467368155253357e-07,
|
| 551 |
+
"loss": 8.2663,
|
| 552 |
+
"step": 17000
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 0.7229428024665108,
|
| 556 |
+
"eval_cosine_accuracy": 0.9686437845230103,
|
| 557 |
+
"eval_loss": 8.462857246398926,
|
| 558 |
+
"eval_runtime": 49.5482,
|
| 559 |
+
"eval_samples_per_second": 381.608,
|
| 560 |
+
"eval_steps_per_second": 1.493,
|
| 561 |
+
"step": 17000
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 0.744205826068467,
|
| 565 |
+
"grad_norm": 14.019209861755371,
|
| 566 |
+
"learning_rate": 8.716056601955618e-07,
|
| 567 |
+
"loss": 8.2225,
|
| 568 |
+
"step": 17500
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 0.744205826068467,
|
| 572 |
+
"eval_cosine_accuracy": 0.9691875576972961,
|
| 573 |
+
"eval_loss": 8.401082038879395,
|
| 574 |
+
"eval_runtime": 50.4644,
|
| 575 |
+
"eval_samples_per_second": 374.68,
|
| 576 |
+
"eval_steps_per_second": 1.466,
|
| 577 |
+
"step": 17500
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.7654688496704232,
|
| 581 |
+
"grad_norm": 13.743831634521484,
|
| 582 |
+
"learning_rate": 8.965243422298566e-07,
|
| 583 |
+
"loss": 8.2052,
|
| 584 |
+
"step": 18000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.7654688496704232,
|
| 588 |
+
"eval_cosine_accuracy": 0.9689610004425049,
|
| 589 |
+
"eval_loss": 8.454690933227539,
|
| 590 |
+
"eval_runtime": 49.7999,
|
| 591 |
+
"eval_samples_per_second": 379.679,
|
| 592 |
+
"eval_steps_per_second": 1.486,
|
| 593 |
+
"step": 18000
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 0.7867318732723794,
|
| 597 |
+
"grad_norm": 14.726578712463379,
|
| 598 |
+
"learning_rate": 9.214430242641513e-07,
|
| 599 |
+
"loss": 8.1894,
|
| 600 |
+
"step": 18500
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 0.7867318732723794,
|
| 604 |
+
"eval_cosine_accuracy": 0.9687797427177429,
|
| 605 |
+
"eval_loss": 8.442588806152344,
|
| 606 |
+
"eval_runtime": 50.3938,
|
| 607 |
+
"eval_samples_per_second": 375.205,
|
| 608 |
+
"eval_steps_per_second": 1.468,
|
| 609 |
+
"step": 18500
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.8079948968743356,
|
| 613 |
+
"grad_norm": 16.874422073364258,
|
| 614 |
+
"learning_rate": 9.463617062984462e-07,
|
| 615 |
+
"loss": 8.1808,
|
| 616 |
+
"step": 19000
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 0.8079948968743356,
|
| 620 |
+
"eval_cosine_accuracy": 0.9687344431877136,
|
| 621 |
+
"eval_loss": 8.419596672058105,
|
| 622 |
+
"eval_runtime": 49.7216,
|
| 623 |
+
"eval_samples_per_second": 380.277,
|
| 624 |
+
"eval_steps_per_second": 1.488,
|
| 625 |
+
"step": 19000
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.8292579204762918,
|
| 629 |
+
"grad_norm": 18.23941993713379,
|
| 630 |
+
"learning_rate": 9.71280388332741e-07,
|
| 631 |
+
"loss": 8.1298,
|
| 632 |
+
"step": 19500
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.8292579204762918,
|
| 636 |
+
"eval_cosine_accuracy": 0.969504714012146,
|
| 637 |
+
"eval_loss": 8.426375389099121,
|
| 638 |
+
"eval_runtime": 50.209,
|
| 639 |
+
"eval_samples_per_second": 376.586,
|
| 640 |
+
"eval_steps_per_second": 1.474,
|
| 641 |
+
"step": 19500
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.850520944078248,
|
| 645 |
+
"grad_norm": 18.061824798583984,
|
| 646 |
+
"learning_rate": 9.961990703670356e-07,
|
| 647 |
+
"loss": 8.1187,
|
| 648 |
+
"step": 20000
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 0.850520944078248,
|
| 652 |
+
"eval_cosine_accuracy": 0.969504714012146,
|
| 653 |
+
"eval_loss": 8.403668403625488,
|
| 654 |
+
"eval_runtime": 50.0426,
|
| 655 |
+
"eval_samples_per_second": 377.838,
|
| 656 |
+
"eval_steps_per_second": 1.479,
|
| 657 |
+
"step": 20000
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 0.8717839676802042,
|
| 661 |
+
"grad_norm": 14.699914932250977,
|
| 662 |
+
"learning_rate": 1.0211177524013304e-06,
|
| 663 |
+
"loss": 8.0909,
|
| 664 |
+
"step": 20500
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 0.8717839676802042,
|
| 668 |
+
"eval_cosine_accuracy": 0.9694141149520874,
|
| 669 |
+
"eval_loss": 8.39807415008545,
|
| 670 |
+
"eval_runtime": 50.3586,
|
| 671 |
+
"eval_samples_per_second": 375.468,
|
| 672 |
+
"eval_steps_per_second": 1.469,
|
| 673 |
+
"step": 20500
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 0.8930469912821604,
|
| 677 |
+
"grad_norm": 31.298362731933594,
|
| 678 |
+
"learning_rate": 1.046036434435625e-06,
|
| 679 |
+
"loss": 8.0338,
|
| 680 |
+
"step": 21000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.8930469912821604,
|
| 684 |
+
"eval_cosine_accuracy": 0.9687344431877136,
|
| 685 |
+
"eval_loss": 8.437390327453613,
|
| 686 |
+
"eval_runtime": 50.0486,
|
| 687 |
+
"eval_samples_per_second": 377.793,
|
| 688 |
+
"eval_steps_per_second": 1.479,
|
| 689 |
+
"step": 21000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.9143100148841166,
|
| 693 |
+
"grad_norm": 20.37915802001953,
|
| 694 |
+
"learning_rate": 1.0709052791058514e-06,
|
| 695 |
+
"loss": 8.0198,
|
| 696 |
+
"step": 21500
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.9143100148841166,
|
| 700 |
+
"eval_cosine_accuracy": 0.9688250422477722,
|
| 701 |
+
"eval_loss": 8.418041229248047,
|
| 702 |
+
"eval_runtime": 50.3563,
|
| 703 |
+
"eval_samples_per_second": 375.484,
|
| 704 |
+
"eval_steps_per_second": 1.47,
|
| 705 |
+
"step": 21500
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 0.9355730384860728,
|
| 709 |
+
"grad_norm": 16.509695053100586,
|
| 710 |
+
"learning_rate": 1.095823961140146e-06,
|
| 711 |
+
"loss": 8.0113,
|
| 712 |
+
"step": 22000
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 0.9355730384860728,
|
| 716 |
+
"eval_cosine_accuracy": 0.968598484992981,
|
| 717 |
+
"eval_loss": 8.439444541931152,
|
| 718 |
+
"eval_runtime": 49.8709,
|
| 719 |
+
"eval_samples_per_second": 379.139,
|
| 720 |
+
"eval_steps_per_second": 1.484,
|
| 721 |
+
"step": 22000
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 0.956836062088029,
|
| 725 |
+
"grad_norm": 18.794130325317383,
|
| 726 |
+
"learning_rate": 1.1207426431744406e-06,
|
| 727 |
+
"loss": 8.0024,
|
| 728 |
+
"step": 22500
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 0.956836062088029,
|
| 732 |
+
"eval_cosine_accuracy": 0.9693688154220581,
|
| 733 |
+
"eval_loss": 8.415027618408203,
|
| 734 |
+
"eval_runtime": 50.599,
|
| 735 |
+
"eval_samples_per_second": 373.683,
|
| 736 |
+
"eval_steps_per_second": 1.462,
|
| 737 |
+
"step": 22500
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.9780990856899852,
|
| 741 |
+
"grad_norm": 26.394250869750977,
|
| 742 |
+
"learning_rate": 1.1456613252087355e-06,
|
| 743 |
+
"loss": 7.9449,
|
| 744 |
+
"step": 23000
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.9780990856899852,
|
| 748 |
+
"eval_cosine_accuracy": 0.9687797427177429,
|
| 749 |
+
"eval_loss": 8.452112197875977,
|
| 750 |
+
"eval_runtime": 49.7524,
|
| 751 |
+
"eval_samples_per_second": 380.042,
|
| 752 |
+
"eval_steps_per_second": 1.487,
|
| 753 |
+
"step": 23000
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.9993621092919414,
|
| 757 |
+
"grad_norm": 17.094562530517578,
|
| 758 |
+
"learning_rate": 1.1705301698789618e-06,
|
| 759 |
+
"loss": 7.9241,
|
| 760 |
+
"step": 23500
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 0.9993621092919414,
|
| 764 |
+
"eval_cosine_accuracy": 0.968598484992981,
|
| 765 |
+
"eval_loss": 8.457128524780273,
|
| 766 |
+
"eval_runtime": 49.272,
|
| 767 |
+
"eval_samples_per_second": 383.747,
|
| 768 |
+
"eval_steps_per_second": 1.502,
|
| 769 |
+
"step": 23500
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 1.0206251328938976,
|
| 773 |
+
"grad_norm": 19.156238555908203,
|
| 774 |
+
"learning_rate": 1.1954488519132564e-06,
|
| 775 |
+
"loss": 7.8644,
|
| 776 |
+
"step": 24000
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 1.0206251328938976,
|
| 780 |
+
"eval_cosine_accuracy": 0.9685531854629517,
|
| 781 |
+
"eval_loss": 8.447549819946289,
|
| 782 |
+
"eval_runtime": 51.0348,
|
| 783 |
+
"eval_samples_per_second": 370.493,
|
| 784 |
+
"eval_steps_per_second": 1.45,
|
| 785 |
+
"step": 24000
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 1.0418881564958538,
|
| 789 |
+
"grad_norm": 17.439102172851562,
|
| 790 |
+
"learning_rate": 1.220367533947551e-06,
|
| 791 |
+
"loss": 7.9011,
|
| 792 |
+
"step": 24500
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 1.0418881564958538,
|
| 796 |
+
"eval_cosine_accuracy": 0.9687797427177429,
|
| 797 |
+
"eval_loss": 8.469432830810547,
|
| 798 |
+
"eval_runtime": 50.4242,
|
| 799 |
+
"eval_samples_per_second": 374.978,
|
| 800 |
+
"eval_steps_per_second": 1.468,
|
| 801 |
+
"step": 24500
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 1.06315118009781,
|
| 805 |
+
"grad_norm": 21.96286964416504,
|
| 806 |
+
"learning_rate": 1.2452862159818461e-06,
|
| 807 |
+
"loss": 7.8623,
|
| 808 |
+
"step": 25000
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 1.06315118009781,
|
| 812 |
+
"eval_cosine_accuracy": 0.9688250422477722,
|
| 813 |
+
"eval_loss": 8.477595329284668,
|
| 814 |
+
"eval_runtime": 49.8803,
|
| 815 |
+
"eval_samples_per_second": 379.068,
|
| 816 |
+
"eval_steps_per_second": 1.484,
|
| 817 |
+
"step": 25000
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 1.0844142036997662,
|
| 821 |
+
"grad_norm": 17.020614624023438,
|
| 822 |
+
"learning_rate": 1.2701550606520722e-06,
|
| 823 |
+
"loss": 7.8451,
|
| 824 |
+
"step": 25500
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 1.0844142036997662,
|
| 828 |
+
"eval_cosine_accuracy": 0.9689157009124756,
|
| 829 |
+
"eval_loss": 8.455941200256348,
|
| 830 |
+
"eval_runtime": 49.3477,
|
| 831 |
+
"eval_samples_per_second": 383.159,
|
| 832 |
+
"eval_steps_per_second": 1.5,
|
| 833 |
+
"step": 25500
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 1.1056772273017224,
|
| 837 |
+
"grad_norm": 16.207447052001953,
|
| 838 |
+
"learning_rate": 1.2950737426863668e-06,
|
| 839 |
+
"loss": 7.7974,
|
| 840 |
+
"step": 26000
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 1.1056772273017224,
|
| 844 |
+
"eval_cosine_accuracy": 0.9693688154220581,
|
| 845 |
+
"eval_loss": 8.481649398803711,
|
| 846 |
+
"eval_runtime": 50.7649,
|
| 847 |
+
"eval_samples_per_second": 372.462,
|
| 848 |
+
"eval_steps_per_second": 1.458,
|
| 849 |
+
"step": 26000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.1269402509036786,
|
| 853 |
+
"grad_norm": 18.933929443359375,
|
| 854 |
+
"learning_rate": 1.3199924247206615e-06,
|
| 855 |
+
"loss": 7.8167,
|
| 856 |
+
"step": 26500
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.1269402509036786,
|
| 860 |
+
"eval_cosine_accuracy": 0.9689157009124756,
|
| 861 |
+
"eval_loss": 8.489095687866211,
|
| 862 |
+
"eval_runtime": 50.3783,
|
| 863 |
+
"eval_samples_per_second": 375.321,
|
| 864 |
+
"eval_steps_per_second": 1.469,
|
| 865 |
+
"step": 26500
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 1.1482032745056348,
|
| 869 |
+
"grad_norm": 21.68393898010254,
|
| 870 |
+
"learning_rate": 1.3449111067549565e-06,
|
| 871 |
+
"loss": 7.7871,
|
| 872 |
+
"step": 27000
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 1.1482032745056348,
|
| 876 |
+
"eval_cosine_accuracy": 0.9683719277381897,
|
| 877 |
+
"eval_loss": 8.516972541809082,
|
| 878 |
+
"eval_runtime": 49.7949,
|
| 879 |
+
"eval_samples_per_second": 379.718,
|
| 880 |
+
"eval_steps_per_second": 1.486,
|
| 881 |
+
"step": 27000
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 1.169466298107591,
|
| 885 |
+
"grad_norm": 29.102264404296875,
|
| 886 |
+
"learning_rate": 1.3697799514251824e-06,
|
| 887 |
+
"loss": 7.7629,
|
| 888 |
+
"step": 27500
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 1.169466298107591,
|
| 892 |
+
"eval_cosine_accuracy": 0.9687797427177429,
|
| 893 |
+
"eval_loss": 8.527128219604492,
|
| 894 |
+
"eval_runtime": 50.4479,
|
| 895 |
+
"eval_samples_per_second": 374.802,
|
| 896 |
+
"eval_steps_per_second": 1.467,
|
| 897 |
+
"step": 27500
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 1.1907293217095472,
|
| 901 |
+
"grad_norm": 29.683128356933594,
|
| 902 |
+
"learning_rate": 1.3946487960954087e-06,
|
| 903 |
+
"loss": 7.7362,
|
| 904 |
+
"step": 28000
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 1.1907293217095472,
|
| 908 |
+
"eval_cosine_accuracy": 0.9690062999725342,
|
| 909 |
+
"eval_loss": 8.560052871704102,
|
| 910 |
+
"eval_runtime": 49.7688,
|
| 911 |
+
"eval_samples_per_second": 379.917,
|
| 912 |
+
"eval_steps_per_second": 1.487,
|
| 913 |
+
"step": 28000
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.2119923453115033,
|
| 917 |
+
"grad_norm": 19.49919891357422,
|
| 918 |
+
"learning_rate": 1.4195674781297035e-06,
|
| 919 |
+
"loss": 7.7203,
|
| 920 |
+
"step": 28500
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.2119923453115033,
|
| 924 |
+
"eval_cosine_accuracy": 0.9681906700134277,
|
| 925 |
+
"eval_loss": 8.618322372436523,
|
| 926 |
+
"eval_runtime": 50.5207,
|
| 927 |
+
"eval_samples_per_second": 374.263,
|
| 928 |
+
"eval_steps_per_second": 1.465,
|
| 929 |
+
"step": 28500
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 1.2332553689134595,
|
| 933 |
+
"grad_norm": 30.64069366455078,
|
| 934 |
+
"learning_rate": 1.4444861601639982e-06,
|
| 935 |
+
"loss": 7.713,
|
| 936 |
+
"step": 29000
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 1.2332553689134595,
|
| 940 |
+
"eval_cosine_accuracy": 0.969504714012146,
|
| 941 |
+
"eval_loss": 8.546183586120605,
|
| 942 |
+
"eval_runtime": 49.6131,
|
| 943 |
+
"eval_samples_per_second": 381.109,
|
| 944 |
+
"eval_steps_per_second": 1.492,
|
| 945 |
+
"step": 29000
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 1.2545183925154157,
|
| 949 |
+
"grad_norm": 21.36481285095215,
|
| 950 |
+
"learning_rate": 1.4694048421982928e-06,
|
| 951 |
+
"loss": 7.6853,
|
| 952 |
+
"step": 29500
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 1.2545183925154157,
|
| 956 |
+
"eval_cosine_accuracy": 0.969323456287384,
|
| 957 |
+
"eval_loss": 8.578022956848145,
|
| 958 |
+
"eval_runtime": 49.7468,
|
| 959 |
+
"eval_samples_per_second": 380.085,
|
| 960 |
+
"eval_steps_per_second": 1.488,
|
| 961 |
+
"step": 29500
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 1.275781416117372,
|
| 965 |
+
"grad_norm": 37.677242279052734,
|
| 966 |
+
"learning_rate": 1.4943235242325879e-06,
|
| 967 |
+
"loss": 7.6376,
|
| 968 |
+
"step": 30000
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 1.275781416117372,
|
| 972 |
+
"eval_cosine_accuracy": 0.9687797427177429,
|
| 973 |
+
"eval_loss": 8.658872604370117,
|
| 974 |
+
"eval_runtime": 51.0348,
|
| 975 |
+
"eval_samples_per_second": 370.492,
|
| 976 |
+
"eval_steps_per_second": 1.45,
|
| 977 |
+
"step": 30000
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 1.2970444397193281,
|
| 981 |
+
"grad_norm": 20.2037296295166,
|
| 982 |
+
"learning_rate": 1.519192368902814e-06,
|
| 983 |
+
"loss": 7.6364,
|
| 984 |
+
"step": 30500
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 1.2970444397193281,
|
| 988 |
+
"eval_cosine_accuracy": 0.9686437845230103,
|
| 989 |
+
"eval_loss": 8.601694107055664,
|
| 990 |
+
"eval_runtime": 50.5448,
|
| 991 |
+
"eval_samples_per_second": 374.084,
|
| 992 |
+
"eval_steps_per_second": 1.464,
|
| 993 |
+
"step": 30500
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 1.3183074633212843,
|
| 997 |
+
"grad_norm": 18.782926559448242,
|
| 998 |
+
"learning_rate": 1.5441110509371086e-06,
|
| 999 |
+
"loss": 7.6196,
|
| 1000 |
+
"step": 31000
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 1.3183074633212843,
|
| 1004 |
+
"eval_cosine_accuracy": 0.968417227268219,
|
| 1005 |
+
"eval_loss": 8.666234970092773,
|
| 1006 |
+
"eval_runtime": 49.549,
|
| 1007 |
+
"eval_samples_per_second": 381.602,
|
| 1008 |
+
"eval_steps_per_second": 1.493,
|
| 1009 |
+
"step": 31000
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 1.3395704869232405,
|
| 1013 |
+
"grad_norm": 24.106821060180664,
|
| 1014 |
+
"learning_rate": 1.5690297329714032e-06,
|
| 1015 |
+
"loss": 7.6169,
|
| 1016 |
+
"step": 31500
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 1.3395704869232405,
|
| 1020 |
+
"eval_cosine_accuracy": 0.9685531854629517,
|
| 1021 |
+
"eval_loss": 8.674742698669434,
|
| 1022 |
+
"eval_runtime": 50.4812,
|
| 1023 |
+
"eval_samples_per_second": 374.555,
|
| 1024 |
+
"eval_steps_per_second": 1.466,
|
| 1025 |
+
"step": 31500
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.3608335105251967,
|
| 1029 |
+
"grad_norm": 21.349994659423828,
|
| 1030 |
+
"learning_rate": 1.5939484150056983e-06,
|
| 1031 |
+
"loss": 7.5608,
|
| 1032 |
+
"step": 32000
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.3608335105251967,
|
| 1036 |
+
"eval_cosine_accuracy": 0.9680547118186951,
|
| 1037 |
+
"eval_loss": 8.65296459197998,
|
| 1038 |
+
"eval_runtime": 49.9031,
|
| 1039 |
+
"eval_samples_per_second": 378.894,
|
| 1040 |
+
"eval_steps_per_second": 1.483,
|
| 1041 |
+
"step": 32000
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 1.382096534127153,
|
| 1045 |
+
"grad_norm": 23.57134246826172,
|
| 1046 |
+
"learning_rate": 1.618867097039993e-06,
|
| 1047 |
+
"loss": 7.5725,
|
| 1048 |
+
"step": 32500
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 1.382096534127153,
|
| 1052 |
+
"eval_cosine_accuracy": 0.9688250422477722,
|
| 1053 |
+
"eval_loss": 8.634788513183594,
|
| 1054 |
+
"eval_runtime": 50.319,
|
| 1055 |
+
"eval_samples_per_second": 375.763,
|
| 1056 |
+
"eval_steps_per_second": 1.471,
|
| 1057 |
+
"step": 32500
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 1.4033595577291091,
|
| 1061 |
+
"grad_norm": 41.65155792236328,
|
| 1062 |
+
"learning_rate": 1.643735941710219e-06,
|
| 1063 |
+
"loss": 7.5056,
|
| 1064 |
+
"step": 33000
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 1.4033595577291091,
|
| 1068 |
+
"eval_cosine_accuracy": 0.967510998249054,
|
| 1069 |
+
"eval_loss": 8.732030868530273,
|
| 1070 |
+
"eval_runtime": 49.7065,
|
| 1071 |
+
"eval_samples_per_second": 380.393,
|
| 1072 |
+
"eval_steps_per_second": 1.489,
|
| 1073 |
+
"step": 33000
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 1.4246225813310653,
|
| 1077 |
+
"grad_norm": 21.274824142456055,
|
| 1078 |
+
"learning_rate": 1.6686546237445138e-06,
|
| 1079 |
+
"loss": 7.5294,
|
| 1080 |
+
"step": 33500
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 1.4246225813310653,
|
| 1084 |
+
"eval_cosine_accuracy": 0.9690968990325928,
|
| 1085 |
+
"eval_loss": 8.727688789367676,
|
| 1086 |
+
"eval_runtime": 50.402,
|
| 1087 |
+
"eval_samples_per_second": 375.144,
|
| 1088 |
+
"eval_steps_per_second": 1.468,
|
| 1089 |
+
"step": 33500
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 1.4458856049330215,
|
| 1093 |
+
"grad_norm": 24.373477935791016,
|
| 1094 |
+
"learning_rate": 1.6935733057788085e-06,
|
| 1095 |
+
"loss": 7.5106,
|
| 1096 |
+
"step": 34000
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 1.4458856049330215,
|
| 1100 |
+
"eval_cosine_accuracy": 0.9679641127586365,
|
| 1101 |
+
"eval_loss": 8.750975608825684,
|
| 1102 |
+
"eval_runtime": 49.6654,
|
| 1103 |
+
"eval_samples_per_second": 380.708,
|
| 1104 |
+
"eval_steps_per_second": 1.49,
|
| 1105 |
+
"step": 34000
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 1.4671486285349777,
|
| 1109 |
+
"grad_norm": 29.188819885253906,
|
| 1110 |
+
"learning_rate": 1.7184919878131033e-06,
|
| 1111 |
+
"loss": 7.5032,
|
| 1112 |
+
"step": 34500
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 1.4671486285349777,
|
| 1116 |
+
"eval_cosine_accuracy": 0.967510998249054,
|
| 1117 |
+
"eval_loss": 8.7284574508667,
|
| 1118 |
+
"eval_runtime": 49.6728,
|
| 1119 |
+
"eval_samples_per_second": 380.651,
|
| 1120 |
+
"eval_steps_per_second": 1.49,
|
| 1121 |
+
"step": 34500
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 1.488411652136934,
|
| 1125 |
+
"grad_norm": 24.691267013549805,
|
| 1126 |
+
"learning_rate": 1.7433608324833296e-06,
|
| 1127 |
+
"loss": 7.4635,
|
| 1128 |
+
"step": 35000
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 1.488411652136934,
|
| 1132 |
+
"eval_cosine_accuracy": 0.968235969543457,
|
| 1133 |
+
"eval_loss": 8.765801429748535,
|
| 1134 |
+
"eval_runtime": 50.8811,
|
| 1135 |
+
"eval_samples_per_second": 371.611,
|
| 1136 |
+
"eval_steps_per_second": 1.454,
|
| 1137 |
+
"step": 35000
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.5096746757388901,
|
| 1141 |
+
"grad_norm": 50.87013626098633,
|
| 1142 |
+
"learning_rate": 1.7682795145176243e-06,
|
| 1143 |
+
"loss": 7.4503,
|
| 1144 |
+
"step": 35500
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.5096746757388901,
|
| 1148 |
+
"eval_cosine_accuracy": 0.9684625267982483,
|
| 1149 |
+
"eval_loss": 8.72378921508789,
|
| 1150 |
+
"eval_runtime": 49.667,
|
| 1151 |
+
"eval_samples_per_second": 380.695,
|
| 1152 |
+
"eval_steps_per_second": 1.49,
|
| 1153 |
+
"step": 35500
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 1.5309376993408463,
|
| 1157 |
+
"grad_norm": 40.06660842895508,
|
| 1158 |
+
"learning_rate": 1.793198196551919e-06,
|
| 1159 |
+
"loss": 7.4227,
|
| 1160 |
+
"step": 36000
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 1.5309376993408463,
|
| 1164 |
+
"eval_cosine_accuracy": 0.9674656987190247,
|
| 1165 |
+
"eval_loss": 8.823667526245117,
|
| 1166 |
+
"eval_runtime": 51.0542,
|
| 1167 |
+
"eval_samples_per_second": 370.352,
|
| 1168 |
+
"eval_steps_per_second": 1.449,
|
| 1169 |
+
"step": 36000
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 1.5522007229428025,
|
| 1173 |
+
"grad_norm": 22.465124130249023,
|
| 1174 |
+
"learning_rate": 1.8181168785862137e-06,
|
| 1175 |
+
"loss": 7.4243,
|
| 1176 |
+
"step": 36500
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 1.5522007229428025,
|
| 1180 |
+
"eval_cosine_accuracy": 0.9679641127586365,
|
| 1181 |
+
"eval_loss": 8.741899490356445,
|
| 1182 |
+
"eval_runtime": 49.6301,
|
| 1183 |
+
"eval_samples_per_second": 380.978,
|
| 1184 |
+
"eval_steps_per_second": 1.491,
|
| 1185 |
+
"step": 36500
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 1.5734637465447587,
|
| 1189 |
+
"grad_norm": 33.76041793823242,
|
| 1190 |
+
"learning_rate": 1.8430355606205084e-06,
|
| 1191 |
+
"loss": 7.3642,
|
| 1192 |
+
"step": 37000
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 1.5734637465447587,
|
| 1196 |
+
"eval_cosine_accuracy": 0.9678281545639038,
|
| 1197 |
+
"eval_loss": 8.775369644165039,
|
| 1198 |
+
"eval_runtime": 50.8941,
|
| 1199 |
+
"eval_samples_per_second": 371.517,
|
| 1200 |
+
"eval_steps_per_second": 1.454,
|
| 1201 |
+
"step": 37000
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 1.594726770146715,
|
| 1205 |
+
"grad_norm": 70.6234359741211,
|
| 1206 |
+
"learning_rate": 1.8679044052907347e-06,
|
| 1207 |
+
"loss": 7.3529,
|
| 1208 |
+
"step": 37500
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 1.594726770146715,
|
| 1212 |
+
"eval_cosine_accuracy": 0.9670125246047974,
|
| 1213 |
+
"eval_loss": 8.834284782409668,
|
| 1214 |
+
"eval_runtime": 49.7944,
|
| 1215 |
+
"eval_samples_per_second": 379.721,
|
| 1216 |
+
"eval_steps_per_second": 1.486,
|
| 1217 |
+
"step": 37500
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 1.6159897937486711,
|
| 1221 |
+
"grad_norm": 20.19392967224121,
|
| 1222 |
+
"learning_rate": 1.8928230873250293e-06,
|
| 1223 |
+
"loss": 7.3594,
|
| 1224 |
+
"step": 38000
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 1.6159897937486711,
|
| 1228 |
+
"eval_cosine_accuracy": 0.9676922559738159,
|
| 1229 |
+
"eval_loss": 8.824420928955078,
|
| 1230 |
+
"eval_runtime": 51.008,
|
| 1231 |
+
"eval_samples_per_second": 370.687,
|
| 1232 |
+
"eval_steps_per_second": 1.451,
|
| 1233 |
+
"step": 38000
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 1.6372528173506273,
|
| 1237 |
+
"grad_norm": 28.939029693603516,
|
| 1238 |
+
"learning_rate": 1.917741769359324e-06,
|
| 1239 |
+
"loss": 7.3257,
|
| 1240 |
+
"step": 38500
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 1.6372528173506273,
|
| 1244 |
+
"eval_cosine_accuracy": 0.9672391414642334,
|
| 1245 |
+
"eval_loss": 8.849225044250488,
|
| 1246 |
+
"eval_runtime": 49.3783,
|
| 1247 |
+
"eval_samples_per_second": 382.921,
|
| 1248 |
+
"eval_steps_per_second": 1.499,
|
| 1249 |
+
"step": 38500
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.6585158409525835,
|
| 1253 |
+
"grad_norm": 24.4239559173584,
|
| 1254 |
+
"learning_rate": 1.9426604513936186e-06,
|
| 1255 |
+
"loss": 7.3277,
|
| 1256 |
+
"step": 39000
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.6585158409525835,
|
| 1260 |
+
"eval_cosine_accuracy": 0.9667859673500061,
|
| 1261 |
+
"eval_loss": 8.900370597839355,
|
| 1262 |
+
"eval_runtime": 50.6295,
|
| 1263 |
+
"eval_samples_per_second": 373.458,
|
| 1264 |
+
"eval_steps_per_second": 1.462,
|
| 1265 |
+
"step": 39000
|
| 1266 |
+
},
|
| 1267 |
+
{
|
| 1268 |
+
"epoch": 1.6797788645545397,
|
| 1269 |
+
"grad_norm": 27.921985626220703,
|
| 1270 |
+
"learning_rate": 1.9675292960638453e-06,
|
| 1271 |
+
"loss": 7.3037,
|
| 1272 |
+
"step": 39500
|
| 1273 |
+
},
|
| 1274 |
+
{
|
| 1275 |
+
"epoch": 1.6797788645545397,
|
| 1276 |
+
"eval_cosine_accuracy": 0.9670578837394714,
|
| 1277 |
+
"eval_loss": 8.904617309570312,
|
| 1278 |
+
"eval_runtime": 49.7604,
|
| 1279 |
+
"eval_samples_per_second": 379.981,
|
| 1280 |
+
"eval_steps_per_second": 1.487,
|
| 1281 |
+
"step": 39500
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 1.701041888156496,
|
| 1285 |
+
"grad_norm": 26.04640007019043,
|
| 1286 |
+
"learning_rate": 1.9924479780981397e-06,
|
| 1287 |
+
"loss": 7.2672,
|
| 1288 |
+
"step": 40000
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"epoch": 1.701041888156496,
|
| 1292 |
+
"eval_cosine_accuracy": 0.9654719233512878,
|
| 1293 |
+
"eval_loss": 9.0310697555542,
|
| 1294 |
+
"eval_runtime": 50.5286,
|
| 1295 |
+
"eval_samples_per_second": 374.204,
|
| 1296 |
+
"eval_steps_per_second": 1.465,
|
| 1297 |
+
"step": 40000
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 1.7223049117584521,
|
| 1301 |
+
"grad_norm": 27.62725830078125,
|
| 1302 |
+
"learning_rate": 2.0173666601324346e-06,
|
| 1303 |
+
"loss": 7.2515,
|
| 1304 |
+
"step": 40500
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 1.7223049117584521,
|
| 1308 |
+
"eval_cosine_accuracy": 0.9665141105651855,
|
| 1309 |
+
"eval_loss": 8.956608772277832,
|
| 1310 |
+
"eval_runtime": 49.684,
|
| 1311 |
+
"eval_samples_per_second": 380.565,
|
| 1312 |
+
"eval_steps_per_second": 1.489,
|
| 1313 |
+
"step": 40500
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 1.7435679353604083,
|
| 1317 |
+
"grad_norm": 24.697010040283203,
|
| 1318 |
+
"learning_rate": 2.0422853421667294e-06,
|
| 1319 |
+
"loss": 7.2205,
|
| 1320 |
+
"step": 41000
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 1.7435679353604083,
|
| 1324 |
+
"eval_cosine_accuracy": 0.96714848279953,
|
| 1325 |
+
"eval_loss": 8.937081336975098,
|
| 1326 |
+
"eval_runtime": 50.465,
|
| 1327 |
+
"eval_samples_per_second": 374.676,
|
| 1328 |
+
"eval_steps_per_second": 1.466,
|
| 1329 |
+
"step": 41000
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 1.7648309589623645,
|
| 1333 |
+
"grad_norm": 29.04935073852539,
|
| 1334 |
+
"learning_rate": 2.0671541868369553e-06,
|
| 1335 |
+
"loss": 7.2263,
|
| 1336 |
+
"step": 41500
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 1.7648309589623645,
|
| 1340 |
+
"eval_cosine_accuracy": 0.9666500687599182,
|
| 1341 |
+
"eval_loss": 8.94690227508545,
|
| 1342 |
+
"eval_runtime": 49.547,
|
| 1343 |
+
"eval_samples_per_second": 381.618,
|
| 1344 |
+
"eval_steps_per_second": 1.494,
|
| 1345 |
+
"step": 41500
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 1.7860939825643207,
|
| 1349 |
+
"grad_norm": 35.25505065917969,
|
| 1350 |
+
"learning_rate": 2.0920230315071816e-06,
|
| 1351 |
+
"loss": 7.273,
|
| 1352 |
+
"step": 42000
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 1.7860939825643207,
|
| 1356 |
+
"eval_cosine_accuracy": 0.9672844409942627,
|
| 1357 |
+
"eval_loss": 8.9337797164917,
|
| 1358 |
+
"eval_runtime": 50.1569,
|
| 1359 |
+
"eval_samples_per_second": 376.977,
|
| 1360 |
+
"eval_steps_per_second": 1.475,
|
| 1361 |
+
"step": 42000
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.807357006166277,
|
| 1365 |
+
"grad_norm": 26.70579719543457,
|
| 1366 |
+
"learning_rate": 2.116941713541476e-06,
|
| 1367 |
+
"loss": 7.1962,
|
| 1368 |
+
"step": 42500
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.807357006166277,
|
| 1372 |
+
"eval_cosine_accuracy": 0.9669219255447388,
|
| 1373 |
+
"eval_loss": 8.93774700164795,
|
| 1374 |
+
"eval_runtime": 49.8395,
|
| 1375 |
+
"eval_samples_per_second": 379.378,
|
| 1376 |
+
"eval_steps_per_second": 1.485,
|
| 1377 |
+
"step": 42500
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 1.8286200297682331,
|
| 1381 |
+
"grad_norm": 25.269804000854492,
|
| 1382 |
+
"learning_rate": 2.1418603955757713e-06,
|
| 1383 |
+
"loss": 7.1955,
|
| 1384 |
+
"step": 43000
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 1.8286200297682331,
|
| 1388 |
+
"eval_cosine_accuracy": 0.965154767036438,
|
| 1389 |
+
"eval_loss": 9.037887573242188,
|
| 1390 |
+
"eval_runtime": 50.6789,
|
| 1391 |
+
"eval_samples_per_second": 373.094,
|
| 1392 |
+
"eval_steps_per_second": 1.46,
|
| 1393 |
+
"step": 43000
|
| 1394 |
+
},
|
| 1395 |
+
{
|
| 1396 |
+
"epoch": 1.8498830533701893,
|
| 1397 |
+
"grad_norm": 26.97606086730957,
|
| 1398 |
+
"learning_rate": 2.166779077610066e-06,
|
| 1399 |
+
"loss": 7.2015,
|
| 1400 |
+
"step": 43500
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 1.8498830533701893,
|
| 1404 |
+
"eval_cosine_accuracy": 0.9658344388008118,
|
| 1405 |
+
"eval_loss": 9.061223030090332,
|
| 1406 |
+
"eval_runtime": 49.7986,
|
| 1407 |
+
"eval_samples_per_second": 379.689,
|
| 1408 |
+
"eval_steps_per_second": 1.486,
|
| 1409 |
+
"step": 43500
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 1.8711460769721455,
|
| 1413 |
+
"grad_norm": 30.354036331176758,
|
| 1414 |
+
"learning_rate": 2.191647922280292e-06,
|
| 1415 |
+
"loss": 7.1363,
|
| 1416 |
+
"step": 44000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 1.8711460769721455,
|
| 1420 |
+
"eval_cosine_accuracy": 0.963840663433075,
|
| 1421 |
+
"eval_loss": 9.144049644470215,
|
| 1422 |
+
"eval_runtime": 50.374,
|
| 1423 |
+
"eval_samples_per_second": 375.352,
|
| 1424 |
+
"eval_steps_per_second": 1.469,
|
| 1425 |
+
"step": 44000
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 1.8924091005741017,
|
| 1429 |
+
"grad_norm": 34.271522521972656,
|
| 1430 |
+
"learning_rate": 2.216566604314587e-06,
|
| 1431 |
+
"loss": 7.0981,
|
| 1432 |
+
"step": 44500
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 1.8924091005741017,
|
| 1436 |
+
"eval_cosine_accuracy": 0.9659703373908997,
|
| 1437 |
+
"eval_loss": 9.090474128723145,
|
| 1438 |
+
"eval_runtime": 50.0017,
|
| 1439 |
+
"eval_samples_per_second": 378.147,
|
| 1440 |
+
"eval_steps_per_second": 1.48,
|
| 1441 |
+
"step": 44500
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 1.913672124176058,
|
| 1445 |
+
"grad_norm": 26.35101890563965,
|
| 1446 |
+
"learning_rate": 2.2414852863488813e-06,
|
| 1447 |
+
"loss": 7.1226,
|
| 1448 |
+
"step": 45000
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 1.913672124176058,
|
| 1452 |
+
"eval_cosine_accuracy": 0.9629797339439392,
|
| 1453 |
+
"eval_loss": 9.179637908935547,
|
| 1454 |
+
"eval_runtime": 50.6563,
|
| 1455 |
+
"eval_samples_per_second": 373.261,
|
| 1456 |
+
"eval_steps_per_second": 1.461,
|
| 1457 |
+
"step": 45000
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 1.9349351477780141,
|
| 1461 |
+
"grad_norm": 30.242382049560547,
|
| 1462 |
+
"learning_rate": 2.266403968383176e-06,
|
| 1463 |
+
"loss": 7.09,
|
| 1464 |
+
"step": 45500
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 1.9349351477780141,
|
| 1468 |
+
"eval_cosine_accuracy": 0.9639313220977783,
|
| 1469 |
+
"eval_loss": 9.226616859436035,
|
| 1470 |
+
"eval_runtime": 49.6704,
|
| 1471 |
+
"eval_samples_per_second": 380.669,
|
| 1472 |
+
"eval_steps_per_second": 1.49,
|
| 1473 |
+
"step": 45500
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.9561981713799703,
|
| 1477 |
+
"grad_norm": 32.14158248901367,
|
| 1478 |
+
"learning_rate": 2.291322650417471e-06,
|
| 1479 |
+
"loss": 7.0502,
|
| 1480 |
+
"step": 46000
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.9561981713799703,
|
| 1484 |
+
"eval_cosine_accuracy": 0.9655625820159912,
|
| 1485 |
+
"eval_loss": 9.134511947631836,
|
| 1486 |
+
"eval_runtime": 50.3879,
|
| 1487 |
+
"eval_samples_per_second": 375.249,
|
| 1488 |
+
"eval_steps_per_second": 1.469,
|
| 1489 |
+
"step": 46000
|
| 1490 |
+
},
|
| 1491 |
+
{
|
| 1492 |
+
"epoch": 1.9774611949819265,
|
| 1493 |
+
"grad_norm": 34.23884201049805,
|
| 1494 |
+
"learning_rate": 2.3161914950876973e-06,
|
| 1495 |
+
"loss": 7.0414,
|
| 1496 |
+
"step": 46500
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 1.9774611949819265,
|
| 1500 |
+
"eval_cosine_accuracy": 0.9636141061782837,
|
| 1501 |
+
"eval_loss": 9.34079647064209,
|
| 1502 |
+
"eval_runtime": 49.6501,
|
| 1503 |
+
"eval_samples_per_second": 380.825,
|
| 1504 |
+
"eval_steps_per_second": 1.49,
|
| 1505 |
+
"step": 46500
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 1.9987242185838827,
|
| 1509 |
+
"grad_norm": 22.901996612548828,
|
| 1510 |
+
"learning_rate": 2.3410603397579235e-06,
|
| 1511 |
+
"loss": 7.074,
|
| 1512 |
+
"step": 47000
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 1.9987242185838827,
|
| 1516 |
+
"eval_cosine_accuracy": 0.9648375511169434,
|
| 1517 |
+
"eval_loss": 9.222514152526855,
|
| 1518 |
+
"eval_runtime": 50.3934,
|
| 1519 |
+
"eval_samples_per_second": 375.208,
|
| 1520 |
+
"eval_steps_per_second": 1.468,
|
| 1521 |
+
"step": 47000
|
| 1522 |
+
}
|
| 1523 |
+
],
|
| 1524 |
+
"logging_steps": 500,
|
| 1525 |
+
"max_steps": 94060,
|
| 1526 |
+
"num_input_tokens_seen": 0,
|
| 1527 |
+
"num_train_epochs": 4,
|
| 1528 |
+
"save_steps": 500,
|
| 1529 |
+
"stateful_callbacks": {
|
| 1530 |
+
"TrainerControl": {
|
| 1531 |
+
"args": {
|
| 1532 |
+
"should_epoch_stop": false,
|
| 1533 |
+
"should_evaluate": false,
|
| 1534 |
+
"should_log": false,
|
| 1535 |
+
"should_save": true,
|
| 1536 |
+
"should_training_stop": false
|
| 1537 |
+
},
|
| 1538 |
+
"attributes": {}
|
| 1539 |
+
}
|
| 1540 |
+
},
|
| 1541 |
+
"total_flos": 0.0,
|
| 1542 |
+
"train_batch_size": 256,
|
| 1543 |
+
"trial_name": null,
|
| 1544 |
+
"trial_params": null
|
| 1545 |
+
}
|
checkpoint-47030/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c3770870a667cd6f2ea8befc5e2f55e628fb63f8d7e4b0166911fa9d63f89275
|
| 3 |
+
size 5624
|
checkpoint-47030/training_metrics.json
ADDED
|
@@ -0,0 +1,570 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metrics_history": [
|
| 3 |
+
{
|
| 4 |
+
"epoch": 0.0212630236019562,
|
| 5 |
+
"global_step": 500,
|
| 6 |
+
"eval_loss": 11.834263801574707,
|
| 7 |
+
"eval_cosine_accuracy": 0.9308079481124878
|
| 8 |
+
},
|
| 9 |
+
{
|
| 10 |
+
"epoch": 0.0425260472039124,
|
| 11 |
+
"global_step": 1000,
|
| 12 |
+
"eval_loss": 11.764849662780762,
|
| 13 |
+
"eval_cosine_accuracy": 0.9313969612121582
|
| 14 |
+
},
|
| 15 |
+
{
|
| 16 |
+
"epoch": 0.0637890708058686,
|
| 17 |
+
"global_step": 1500,
|
| 18 |
+
"eval_loss": 11.667285919189453,
|
| 19 |
+
"eval_cosine_accuracy": 0.9325751066207886
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"epoch": 0.0850520944078248,
|
| 23 |
+
"global_step": 2000,
|
| 24 |
+
"eval_loss": 11.543947219848633,
|
| 25 |
+
"eval_cosine_accuracy": 0.934478223323822
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.106315118009781,
|
| 29 |
+
"global_step": 2500,
|
| 30 |
+
"eval_loss": 11.422866821289062,
|
| 31 |
+
"eval_cosine_accuracy": 0.9360641837120056
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.1275781416117372,
|
| 35 |
+
"global_step": 3000,
|
| 36 |
+
"eval_loss": 11.268472671508789,
|
| 37 |
+
"eval_cosine_accuracy": 0.9384657144546509
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.1488411652136934,
|
| 41 |
+
"global_step": 3500,
|
| 42 |
+
"eval_loss": 11.027884483337402,
|
| 43 |
+
"eval_cosine_accuracy": 0.941592276096344
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 0.1701041888156496,
|
| 47 |
+
"global_step": 4000,
|
| 48 |
+
"eval_loss": 10.77338981628418,
|
| 49 |
+
"eval_cosine_accuracy": 0.9445375800132751
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.1913672124176058,
|
| 53 |
+
"global_step": 4500,
|
| 54 |
+
"eval_loss": 10.502735137939453,
|
| 55 |
+
"eval_cosine_accuracy": 0.9471657276153564
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.212630236019562,
|
| 59 |
+
"global_step": 5000,
|
| 60 |
+
"eval_loss": 10.223851203918457,
|
| 61 |
+
"eval_cosine_accuracy": 0.9499297738075256
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.2338932596215182,
|
| 65 |
+
"global_step": 5500,
|
| 66 |
+
"eval_loss": 9.879401206970215,
|
| 67 |
+
"eval_cosine_accuracy": 0.9533734917640686
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.2551562832234744,
|
| 71 |
+
"global_step": 6000,
|
| 72 |
+
"eval_loss": 9.583606719970703,
|
| 73 |
+
"eval_cosine_accuracy": 0.9558656811714172
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.2764193068254306,
|
| 77 |
+
"global_step": 6500,
|
| 78 |
+
"eval_loss": 9.295426368713379,
|
| 79 |
+
"eval_cosine_accuracy": 0.9583125710487366
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.2976823304273868,
|
| 83 |
+
"global_step": 7000,
|
| 84 |
+
"eval_loss": 9.070219993591309,
|
| 85 |
+
"eval_cosine_accuracy": 0.9617562890052795
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.318945354029343,
|
| 89 |
+
"global_step": 7500,
|
| 90 |
+
"eval_loss": 8.921951293945312,
|
| 91 |
+
"eval_cosine_accuracy": 0.9646109938621521
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.3402083776312992,
|
| 95 |
+
"global_step": 8000,
|
| 96 |
+
"eval_loss": 8.841903686523438,
|
| 97 |
+
"eval_cosine_accuracy": 0.9665141105651855
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.3614714012332554,
|
| 101 |
+
"global_step": 8500,
|
| 102 |
+
"eval_loss": 8.776097297668457,
|
| 103 |
+
"eval_cosine_accuracy": 0.9681000709533691
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 0.3827344248352116,
|
| 107 |
+
"global_step": 9000,
|
| 108 |
+
"eval_loss": 8.712888717651367,
|
| 109 |
+
"eval_cosine_accuracy": 0.9681000709533691
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.4039974484371678,
|
| 113 |
+
"global_step": 9500,
|
| 114 |
+
"eval_loss": 8.671957015991211,
|
| 115 |
+
"eval_cosine_accuracy": 0.968235969543457
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.425260472039124,
|
| 119 |
+
"global_step": 10000,
|
| 120 |
+
"eval_loss": 8.636693000793457,
|
| 121 |
+
"eval_cosine_accuracy": 0.9679641127586365
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.4465234956410802,
|
| 125 |
+
"global_step": 10500,
|
| 126 |
+
"eval_loss": 8.61970329284668,
|
| 127 |
+
"eval_cosine_accuracy": 0.9684625267982483
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 0.4677865192430364,
|
| 131 |
+
"global_step": 11000,
|
| 132 |
+
"eval_loss": 8.59864616394043,
|
| 133 |
+
"eval_cosine_accuracy": 0.968598484992981
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 0.4890495428449926,
|
| 137 |
+
"global_step": 11500,
|
| 138 |
+
"eval_loss": 8.581836700439453,
|
| 139 |
+
"eval_cosine_accuracy": 0.9678281545639038
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.5103125664469488,
|
| 143 |
+
"global_step": 12000,
|
| 144 |
+
"eval_loss": 8.53200912475586,
|
| 145 |
+
"eval_cosine_accuracy": 0.968598484992981
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.531575590048905,
|
| 149 |
+
"global_step": 12500,
|
| 150 |
+
"eval_loss": 8.522621154785156,
|
| 151 |
+
"eval_cosine_accuracy": 0.9685078859329224
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.5528386136508612,
|
| 155 |
+
"global_step": 13000,
|
| 156 |
+
"eval_loss": 8.52730941772461,
|
| 157 |
+
"eval_cosine_accuracy": 0.9688250422477722
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.5741016372528174,
|
| 161 |
+
"global_step": 13500,
|
| 162 |
+
"eval_loss": 8.488916397094727,
|
| 163 |
+
"eval_cosine_accuracy": 0.9687797427177429
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.5953646608547736,
|
| 167 |
+
"global_step": 14000,
|
| 168 |
+
"eval_loss": 8.479202270507812,
|
| 169 |
+
"eval_cosine_accuracy": 0.9692781567573547
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.6166276844567298,
|
| 173 |
+
"global_step": 14500,
|
| 174 |
+
"eval_loss": 8.463949203491211,
|
| 175 |
+
"eval_cosine_accuracy": 0.9690062999725342
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 0.637890708058686,
|
| 179 |
+
"global_step": 15000,
|
| 180 |
+
"eval_loss": 8.437935829162598,
|
| 181 |
+
"eval_cosine_accuracy": 0.9686437845230103
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.6591537316606422,
|
| 185 |
+
"global_step": 15500,
|
| 186 |
+
"eval_loss": 8.473112106323242,
|
| 187 |
+
"eval_cosine_accuracy": 0.9689610004425049
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 0.6804167552625984,
|
| 191 |
+
"global_step": 16000,
|
| 192 |
+
"eval_loss": 8.445318222045898,
|
| 193 |
+
"eval_cosine_accuracy": 0.9688703417778015
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.7016797788645546,
|
| 197 |
+
"global_step": 16500,
|
| 198 |
+
"eval_loss": 8.421878814697266,
|
| 199 |
+
"eval_cosine_accuracy": 0.9694594144821167
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.7229428024665108,
|
| 203 |
+
"global_step": 17000,
|
| 204 |
+
"eval_loss": 8.462857246398926,
|
| 205 |
+
"eval_cosine_accuracy": 0.9686437845230103
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.744205826068467,
|
| 209 |
+
"global_step": 17500,
|
| 210 |
+
"eval_loss": 8.401082038879395,
|
| 211 |
+
"eval_cosine_accuracy": 0.9691875576972961
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.7654688496704232,
|
| 215 |
+
"global_step": 18000,
|
| 216 |
+
"eval_loss": 8.454690933227539,
|
| 217 |
+
"eval_cosine_accuracy": 0.9689610004425049
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.7867318732723794,
|
| 221 |
+
"global_step": 18500,
|
| 222 |
+
"eval_loss": 8.442588806152344,
|
| 223 |
+
"eval_cosine_accuracy": 0.9687797427177429
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 0.8079948968743356,
|
| 227 |
+
"global_step": 19000,
|
| 228 |
+
"eval_loss": 8.419596672058105,
|
| 229 |
+
"eval_cosine_accuracy": 0.9687344431877136
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 0.8292579204762918,
|
| 233 |
+
"global_step": 19500,
|
| 234 |
+
"eval_loss": 8.426375389099121,
|
| 235 |
+
"eval_cosine_accuracy": 0.969504714012146
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 0.850520944078248,
|
| 239 |
+
"global_step": 20000,
|
| 240 |
+
"eval_loss": 8.403668403625488,
|
| 241 |
+
"eval_cosine_accuracy": 0.969504714012146
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.8717839676802042,
|
| 245 |
+
"global_step": 20500,
|
| 246 |
+
"eval_loss": 8.39807415008545,
|
| 247 |
+
"eval_cosine_accuracy": 0.9694141149520874
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.8930469912821604,
|
| 251 |
+
"global_step": 21000,
|
| 252 |
+
"eval_loss": 8.437390327453613,
|
| 253 |
+
"eval_cosine_accuracy": 0.9687344431877136
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.9143100148841166,
|
| 257 |
+
"global_step": 21500,
|
| 258 |
+
"eval_loss": 8.418041229248047,
|
| 259 |
+
"eval_cosine_accuracy": 0.9688250422477722
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 0.9355730384860728,
|
| 263 |
+
"global_step": 22000,
|
| 264 |
+
"eval_loss": 8.439444541931152,
|
| 265 |
+
"eval_cosine_accuracy": 0.968598484992981
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.956836062088029,
|
| 269 |
+
"global_step": 22500,
|
| 270 |
+
"eval_loss": 8.415027618408203,
|
| 271 |
+
"eval_cosine_accuracy": 0.9693688154220581
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.9780990856899852,
|
| 275 |
+
"global_step": 23000,
|
| 276 |
+
"eval_loss": 8.452112197875977,
|
| 277 |
+
"eval_cosine_accuracy": 0.9687797427177429
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 0.9993621092919414,
|
| 281 |
+
"global_step": 23500,
|
| 282 |
+
"eval_loss": 8.457128524780273,
|
| 283 |
+
"eval_cosine_accuracy": 0.968598484992981
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.0206251328938976,
|
| 287 |
+
"global_step": 24000,
|
| 288 |
+
"eval_loss": 8.447549819946289,
|
| 289 |
+
"eval_cosine_accuracy": 0.9685531854629517
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 1.0418881564958538,
|
| 293 |
+
"global_step": 24500,
|
| 294 |
+
"eval_loss": 8.469432830810547,
|
| 295 |
+
"eval_cosine_accuracy": 0.9687797427177429
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 1.06315118009781,
|
| 299 |
+
"global_step": 25000,
|
| 300 |
+
"eval_loss": 8.477595329284668,
|
| 301 |
+
"eval_cosine_accuracy": 0.9688250422477722
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 1.0844142036997662,
|
| 305 |
+
"global_step": 25500,
|
| 306 |
+
"eval_loss": 8.455941200256348,
|
| 307 |
+
"eval_cosine_accuracy": 0.9689157009124756
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 1.1056772273017224,
|
| 311 |
+
"global_step": 26000,
|
| 312 |
+
"eval_loss": 8.481649398803711,
|
| 313 |
+
"eval_cosine_accuracy": 0.9693688154220581
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 1.1269402509036786,
|
| 317 |
+
"global_step": 26500,
|
| 318 |
+
"eval_loss": 8.489095687866211,
|
| 319 |
+
"eval_cosine_accuracy": 0.9689157009124756
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 1.1482032745056348,
|
| 323 |
+
"global_step": 27000,
|
| 324 |
+
"eval_loss": 8.516972541809082,
|
| 325 |
+
"eval_cosine_accuracy": 0.9683719277381897
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.169466298107591,
|
| 329 |
+
"global_step": 27500,
|
| 330 |
+
"eval_loss": 8.527128219604492,
|
| 331 |
+
"eval_cosine_accuracy": 0.9687797427177429
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.1907293217095472,
|
| 335 |
+
"global_step": 28000,
|
| 336 |
+
"eval_loss": 8.560052871704102,
|
| 337 |
+
"eval_cosine_accuracy": 0.9690062999725342
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 1.2119923453115033,
|
| 341 |
+
"global_step": 28500,
|
| 342 |
+
"eval_loss": 8.618322372436523,
|
| 343 |
+
"eval_cosine_accuracy": 0.9681906700134277
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 1.2332553689134595,
|
| 347 |
+
"global_step": 29000,
|
| 348 |
+
"eval_loss": 8.546183586120605,
|
| 349 |
+
"eval_cosine_accuracy": 0.969504714012146
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 1.2545183925154157,
|
| 353 |
+
"global_step": 29500,
|
| 354 |
+
"eval_loss": 8.578022956848145,
|
| 355 |
+
"eval_cosine_accuracy": 0.969323456287384
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 1.275781416117372,
|
| 359 |
+
"global_step": 30000,
|
| 360 |
+
"eval_loss": 8.658872604370117,
|
| 361 |
+
"eval_cosine_accuracy": 0.9687797427177429
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 1.2970444397193281,
|
| 365 |
+
"global_step": 30500,
|
| 366 |
+
"eval_loss": 8.601694107055664,
|
| 367 |
+
"eval_cosine_accuracy": 0.9686437845230103
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.3183074633212843,
|
| 371 |
+
"global_step": 31000,
|
| 372 |
+
"eval_loss": 8.666234970092773,
|
| 373 |
+
"eval_cosine_accuracy": 0.968417227268219
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 1.3395704869232405,
|
| 377 |
+
"global_step": 31500,
|
| 378 |
+
"eval_loss": 8.674742698669434,
|
| 379 |
+
"eval_cosine_accuracy": 0.9685531854629517
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 1.3608335105251967,
|
| 383 |
+
"global_step": 32000,
|
| 384 |
+
"eval_loss": 8.65296459197998,
|
| 385 |
+
"eval_cosine_accuracy": 0.9680547118186951
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 1.382096534127153,
|
| 389 |
+
"global_step": 32500,
|
| 390 |
+
"eval_loss": 8.634788513183594,
|
| 391 |
+
"eval_cosine_accuracy": 0.9688250422477722
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 1.4033595577291091,
|
| 395 |
+
"global_step": 33000,
|
| 396 |
+
"eval_loss": 8.732030868530273,
|
| 397 |
+
"eval_cosine_accuracy": 0.967510998249054
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 1.4246225813310653,
|
| 401 |
+
"global_step": 33500,
|
| 402 |
+
"eval_loss": 8.727688789367676,
|
| 403 |
+
"eval_cosine_accuracy": 0.9690968990325928
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 1.4458856049330215,
|
| 407 |
+
"global_step": 34000,
|
| 408 |
+
"eval_loss": 8.750975608825684,
|
| 409 |
+
"eval_cosine_accuracy": 0.9679641127586365
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.4671486285349777,
|
| 413 |
+
"global_step": 34500,
|
| 414 |
+
"eval_loss": 8.7284574508667,
|
| 415 |
+
"eval_cosine_accuracy": 0.967510998249054
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.488411652136934,
|
| 419 |
+
"global_step": 35000,
|
| 420 |
+
"eval_loss": 8.765801429748535,
|
| 421 |
+
"eval_cosine_accuracy": 0.968235969543457
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 1.5096746757388901,
|
| 425 |
+
"global_step": 35500,
|
| 426 |
+
"eval_loss": 8.72378921508789,
|
| 427 |
+
"eval_cosine_accuracy": 0.9684625267982483
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 1.5309376993408463,
|
| 431 |
+
"global_step": 36000,
|
| 432 |
+
"eval_loss": 8.823667526245117,
|
| 433 |
+
"eval_cosine_accuracy": 0.9674656987190247
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 1.5522007229428025,
|
| 437 |
+
"global_step": 36500,
|
| 438 |
+
"eval_loss": 8.741899490356445,
|
| 439 |
+
"eval_cosine_accuracy": 0.9679641127586365
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 1.5734637465447587,
|
| 443 |
+
"global_step": 37000,
|
| 444 |
+
"eval_loss": 8.775369644165039,
|
| 445 |
+
"eval_cosine_accuracy": 0.9678281545639038
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 1.594726770146715,
|
| 449 |
+
"global_step": 37500,
|
| 450 |
+
"eval_loss": 8.834284782409668,
|
| 451 |
+
"eval_cosine_accuracy": 0.9670125246047974
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.6159897937486711,
|
| 455 |
+
"global_step": 38000,
|
| 456 |
+
"eval_loss": 8.824420928955078,
|
| 457 |
+
"eval_cosine_accuracy": 0.9676922559738159
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.6372528173506273,
|
| 461 |
+
"global_step": 38500,
|
| 462 |
+
"eval_loss": 8.849225044250488,
|
| 463 |
+
"eval_cosine_accuracy": 0.9672391414642334
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 1.6585158409525835,
|
| 467 |
+
"global_step": 39000,
|
| 468 |
+
"eval_loss": 8.900370597839355,
|
| 469 |
+
"eval_cosine_accuracy": 0.9667859673500061
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 1.6797788645545397,
|
| 473 |
+
"global_step": 39500,
|
| 474 |
+
"eval_loss": 8.904617309570312,
|
| 475 |
+
"eval_cosine_accuracy": 0.9670578837394714
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 1.701041888156496,
|
| 479 |
+
"global_step": 40000,
|
| 480 |
+
"eval_loss": 9.0310697555542,
|
| 481 |
+
"eval_cosine_accuracy": 0.9654719233512878
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 1.7223049117584521,
|
| 485 |
+
"global_step": 40500,
|
| 486 |
+
"eval_loss": 8.956608772277832,
|
| 487 |
+
"eval_cosine_accuracy": 0.9665141105651855
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 1.7435679353604083,
|
| 491 |
+
"global_step": 41000,
|
| 492 |
+
"eval_loss": 8.937081336975098,
|
| 493 |
+
"eval_cosine_accuracy": 0.96714848279953
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 1.7648309589623645,
|
| 497 |
+
"global_step": 41500,
|
| 498 |
+
"eval_loss": 8.94690227508545,
|
| 499 |
+
"eval_cosine_accuracy": 0.9666500687599182
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.7860939825643207,
|
| 503 |
+
"global_step": 42000,
|
| 504 |
+
"eval_loss": 8.9337797164917,
|
| 505 |
+
"eval_cosine_accuracy": 0.9672844409942627
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 1.807357006166277,
|
| 509 |
+
"global_step": 42500,
|
| 510 |
+
"eval_loss": 8.93774700164795,
|
| 511 |
+
"eval_cosine_accuracy": 0.9669219255447388
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 1.8286200297682331,
|
| 515 |
+
"global_step": 43000,
|
| 516 |
+
"eval_loss": 9.037887573242188,
|
| 517 |
+
"eval_cosine_accuracy": 0.965154767036438
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 1.8498830533701893,
|
| 521 |
+
"global_step": 43500,
|
| 522 |
+
"eval_loss": 9.061223030090332,
|
| 523 |
+
"eval_cosine_accuracy": 0.9658344388008118
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 1.8711460769721455,
|
| 527 |
+
"global_step": 44000,
|
| 528 |
+
"eval_loss": 9.144049644470215,
|
| 529 |
+
"eval_cosine_accuracy": 0.963840663433075
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 1.8924091005741017,
|
| 533 |
+
"global_step": 44500,
|
| 534 |
+
"eval_loss": 9.090474128723145,
|
| 535 |
+
"eval_cosine_accuracy": 0.9659703373908997
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 1.913672124176058,
|
| 539 |
+
"global_step": 45000,
|
| 540 |
+
"eval_loss": 9.179637908935547,
|
| 541 |
+
"eval_cosine_accuracy": 0.9629797339439392
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.9349351477780141,
|
| 545 |
+
"global_step": 45500,
|
| 546 |
+
"eval_loss": 9.226616859436035,
|
| 547 |
+
"eval_cosine_accuracy": 0.9639313220977783
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 1.9561981713799703,
|
| 551 |
+
"global_step": 46000,
|
| 552 |
+
"eval_loss": 9.134511947631836,
|
| 553 |
+
"eval_cosine_accuracy": 0.9655625820159912
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 1.9774611949819265,
|
| 557 |
+
"global_step": 46500,
|
| 558 |
+
"eval_loss": 9.34079647064209,
|
| 559 |
+
"eval_cosine_accuracy": 0.9636141061782837
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 1.9987242185838827,
|
| 563 |
+
"global_step": 47000,
|
| 564 |
+
"eval_loss": 9.222514152526855,
|
| 565 |
+
"eval_cosine_accuracy": 0.9648375511169434
|
| 566 |
+
}
|
| 567 |
+
],
|
| 568 |
+
"current_epoch": 2,
|
| 569 |
+
"global_step": 47030
|
| 570 |
+
}
|
checkpoint-47030/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|