Training checkpoint - Epoch 2, Step 57042
Browse files- checkpoint-57042/1_Pooling/config.json +10 -0
- checkpoint-57042/config.json +25 -0
- checkpoint-57042/config_sentence_transformers.json +14 -0
- checkpoint-57042/model.safetensors +3 -0
- checkpoint-57042/modules.json +20 -0
- checkpoint-57042/optimizer.pt +3 -0
- checkpoint-57042/rng_state.pth +3 -0
- checkpoint-57042/scaler.pt +3 -0
- checkpoint-57042/scheduler.pt +3 -0
- checkpoint-57042/sentence_bert_config.json +4 -0
- checkpoint-57042/special_tokens_map.json +37 -0
- checkpoint-57042/tokenizer.json +0 -0
- checkpoint-57042/tokenizer_config.json +65 -0
- checkpoint-57042/trainer_state.json +1865 -0
- checkpoint-57042/training_args.bin +3 -0
- checkpoint-57042/training_metrics.json +690 -0
- checkpoint-57042/vocab.txt +0 -0
checkpoint-57042/1_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-57042/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-57042/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-57042/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7fa4dd5e5ab58a0ce773dec39f759f8ff139c6b50c44beaf16703ec6c795a758
|
| 3 |
+
size 90864192
|
checkpoint-57042/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-57042/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dd99288d94d2f78e5a9a216377148413a9e295d669dec896120b1e7506ef0ad6
|
| 3 |
+
size 180607738
|
checkpoint-57042/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:700fead5c89801a559d151a1cea19ed26b88519f691c5cfd529302092e558fc3
|
| 3 |
+
size 14244
|
checkpoint-57042/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ba0c521ea971aa39030ac11278b6703a0ece08ef82b003502b2c8e0190c8d51c
|
| 3 |
+
size 988
|
checkpoint-57042/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:07d9e8080fec39842ebaec88e69ae07e33a43214c8e33825c558a3c743194bed
|
| 3 |
+
size 1064
|
checkpoint-57042/sentence_bert_config.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"max_seq_length": 256,
|
| 3 |
+
"do_lower_case": false
|
| 4 |
+
}
|
checkpoint-57042/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-57042/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-57042/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-57042/trainer_state.json
ADDED
|
@@ -0,0 +1,1865 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": 57042,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 3.5061884225658285e-05,
|
| 14 |
+
"grad_norm": Infinity,
|
| 15 |
+
"learning_rate": 0.0,
|
| 16 |
+
"loss": 14.1655,
|
| 17 |
+
"step": 1
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.017530942112829143,
|
| 21 |
+
"grad_norm": 46.11793899536133,
|
| 22 |
+
"learning_rate": 8.695652173913044e-07,
|
| 23 |
+
"loss": 15.3746,
|
| 24 |
+
"step": 500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.017530942112829143,
|
| 28 |
+
"eval_cosine_accuracy": 0.9304001331329346,
|
| 29 |
+
"eval_loss": 10.808653831481934,
|
| 30 |
+
"eval_runtime": 50.0101,
|
| 31 |
+
"eval_samples_per_second": 378.244,
|
| 32 |
+
"eval_steps_per_second": 1.48,
|
| 33 |
+
"step": 500
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.035061884225658285,
|
| 37 |
+
"grad_norm": 6.935111045837402,
|
| 38 |
+
"learning_rate": 1.7461430575035066e-06,
|
| 39 |
+
"loss": 11.7216,
|
| 40 |
+
"step": 1000
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.035061884225658285,
|
| 44 |
+
"eval_cosine_accuracy": 0.9219266772270203,
|
| 45 |
+
"eval_loss": 9.035820007324219,
|
| 46 |
+
"eval_runtime": 51.3273,
|
| 47 |
+
"eval_samples_per_second": 368.537,
|
| 48 |
+
"eval_steps_per_second": 1.442,
|
| 49 |
+
"step": 1000
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.05259282633848743,
|
| 53 |
+
"grad_norm": 4.2704758644104,
|
| 54 |
+
"learning_rate": 2.622720897615709e-06,
|
| 55 |
+
"loss": 10.0701,
|
| 56 |
+
"step": 1500
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.05259282633848743,
|
| 60 |
+
"eval_cosine_accuracy": 0.9366985559463501,
|
| 61 |
+
"eval_loss": 8.834137916564941,
|
| 62 |
+
"eval_runtime": 50.3538,
|
| 63 |
+
"eval_samples_per_second": 375.662,
|
| 64 |
+
"eval_steps_per_second": 1.47,
|
| 65 |
+
"step": 1500
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.07012376845131657,
|
| 69 |
+
"grad_norm": 3.6294240951538086,
|
| 70 |
+
"learning_rate": 3.49929873772791e-06,
|
| 71 |
+
"loss": 9.9787,
|
| 72 |
+
"step": 2000
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.07012376845131657,
|
| 76 |
+
"eval_cosine_accuracy": 0.9407313466072083,
|
| 77 |
+
"eval_loss": 8.768850326538086,
|
| 78 |
+
"eval_runtime": 50.9215,
|
| 79 |
+
"eval_samples_per_second": 371.474,
|
| 80 |
+
"eval_steps_per_second": 1.453,
|
| 81 |
+
"step": 2000
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.08765471056414571,
|
| 85 |
+
"grad_norm": 2.800403356552124,
|
| 86 |
+
"learning_rate": 4.375876577840113e-06,
|
| 87 |
+
"loss": 9.9256,
|
| 88 |
+
"step": 2500
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.08765471056414571,
|
| 92 |
+
"eval_cosine_accuracy": 0.9402782320976257,
|
| 93 |
+
"eval_loss": 8.751864433288574,
|
| 94 |
+
"eval_runtime": 50.75,
|
| 95 |
+
"eval_samples_per_second": 372.729,
|
| 96 |
+
"eval_steps_per_second": 1.458,
|
| 97 |
+
"step": 2500
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.10518565267697486,
|
| 101 |
+
"grad_norm": 2.86800217628479,
|
| 102 |
+
"learning_rate": 5.252454417952315e-06,
|
| 103 |
+
"loss": 9.8971,
|
| 104 |
+
"step": 3000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.10518565267697486,
|
| 108 |
+
"eval_cosine_accuracy": 0.9425438642501831,
|
| 109 |
+
"eval_loss": 8.732973098754883,
|
| 110 |
+
"eval_runtime": 51.0941,
|
| 111 |
+
"eval_samples_per_second": 370.219,
|
| 112 |
+
"eval_steps_per_second": 1.448,
|
| 113 |
+
"step": 3000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.122716594789804,
|
| 117 |
+
"grad_norm": 2.924982786178589,
|
| 118 |
+
"learning_rate": 6.129032258064517e-06,
|
| 119 |
+
"loss": 9.8627,
|
| 120 |
+
"step": 3500
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.122716594789804,
|
| 124 |
+
"eval_cosine_accuracy": 0.9468938112258911,
|
| 125 |
+
"eval_loss": 8.691949844360352,
|
| 126 |
+
"eval_runtime": 49.4971,
|
| 127 |
+
"eval_samples_per_second": 382.164,
|
| 128 |
+
"eval_steps_per_second": 1.495,
|
| 129 |
+
"step": 3500
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.14024753690263314,
|
| 133 |
+
"grad_norm": 3.102349281311035,
|
| 134 |
+
"learning_rate": 7.0056100981767185e-06,
|
| 135 |
+
"loss": 9.8413,
|
| 136 |
+
"step": 4000
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.14024753690263314,
|
| 140 |
+
"eval_cosine_accuracy": 0.9477094411849976,
|
| 141 |
+
"eval_loss": 8.692930221557617,
|
| 142 |
+
"eval_runtime": 50.0327,
|
| 143 |
+
"eval_samples_per_second": 378.073,
|
| 144 |
+
"eval_steps_per_second": 1.479,
|
| 145 |
+
"step": 4000
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.1577784790154623,
|
| 149 |
+
"grad_norm": 3.3259987831115723,
|
| 150 |
+
"learning_rate": 7.882187938288921e-06,
|
| 151 |
+
"loss": 9.8187,
|
| 152 |
+
"step": 4500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.1577784790154623,
|
| 156 |
+
"eval_cosine_accuracy": 0.9502469301223755,
|
| 157 |
+
"eval_loss": 8.666604995727539,
|
| 158 |
+
"eval_runtime": 49.5796,
|
| 159 |
+
"eval_samples_per_second": 381.528,
|
| 160 |
+
"eval_steps_per_second": 1.493,
|
| 161 |
+
"step": 4500
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.17530942112829143,
|
| 165 |
+
"grad_norm": 3.461524248123169,
|
| 166 |
+
"learning_rate": 8.758765778401123e-06,
|
| 167 |
+
"loss": 9.797,
|
| 168 |
+
"step": 5000
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 0.17530942112829143,
|
| 172 |
+
"eval_cosine_accuracy": 0.9490235447883606,
|
| 173 |
+
"eval_loss": 8.687457084655762,
|
| 174 |
+
"eval_runtime": 50.1782,
|
| 175 |
+
"eval_samples_per_second": 376.977,
|
| 176 |
+
"eval_steps_per_second": 1.475,
|
| 177 |
+
"step": 5000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.19284036324112058,
|
| 181 |
+
"grad_norm": 3.2629499435424805,
|
| 182 |
+
"learning_rate": 9.635343618513324e-06,
|
| 183 |
+
"loss": 9.7789,
|
| 184 |
+
"step": 5500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.19284036324112058,
|
| 188 |
+
"eval_cosine_accuracy": 0.9515156745910645,
|
| 189 |
+
"eval_loss": 8.682653427124023,
|
| 190 |
+
"eval_runtime": 49.9304,
|
| 191 |
+
"eval_samples_per_second": 378.847,
|
| 192 |
+
"eval_steps_per_second": 1.482,
|
| 193 |
+
"step": 5500
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.2103713053539497,
|
| 197 |
+
"grad_norm": 2.80902361869812,
|
| 198 |
+
"learning_rate": 1.0511921458625528e-05,
|
| 199 |
+
"loss": 9.7601,
|
| 200 |
+
"step": 6000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.2103713053539497,
|
| 204 |
+
"eval_cosine_accuracy": 0.9527844190597534,
|
| 205 |
+
"eval_loss": 8.673870086669922,
|
| 206 |
+
"eval_runtime": 49.5168,
|
| 207 |
+
"eval_samples_per_second": 382.012,
|
| 208 |
+
"eval_steps_per_second": 1.494,
|
| 209 |
+
"step": 6000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.22790224746677887,
|
| 213 |
+
"grad_norm": 3.6036148071289062,
|
| 214 |
+
"learning_rate": 1.138849929873773e-05,
|
| 215 |
+
"loss": 9.7449,
|
| 216 |
+
"step": 6500
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.22790224746677887,
|
| 220 |
+
"eval_cosine_accuracy": 0.9525125622749329,
|
| 221 |
+
"eval_loss": 8.690710067749023,
|
| 222 |
+
"eval_runtime": 51.4523,
|
| 223 |
+
"eval_samples_per_second": 367.641,
|
| 224 |
+
"eval_steps_per_second": 1.438,
|
| 225 |
+
"step": 6500
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.245433189579608,
|
| 229 |
+
"grad_norm": 3.008776903152466,
|
| 230 |
+
"learning_rate": 1.226507713884993e-05,
|
| 231 |
+
"loss": 9.7281,
|
| 232 |
+
"step": 7000
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.245433189579608,
|
| 236 |
+
"eval_cosine_accuracy": 0.9523313045501709,
|
| 237 |
+
"eval_loss": 8.720589637756348,
|
| 238 |
+
"eval_runtime": 50.4169,
|
| 239 |
+
"eval_samples_per_second": 375.192,
|
| 240 |
+
"eval_steps_per_second": 1.468,
|
| 241 |
+
"step": 7000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.26296413169243715,
|
| 245 |
+
"grad_norm": 3.2251088619232178,
|
| 246 |
+
"learning_rate": 1.3141654978962133e-05,
|
| 247 |
+
"loss": 9.7162,
|
| 248 |
+
"step": 7500
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.26296413169243715,
|
| 252 |
+
"eval_cosine_accuracy": 0.9521500468254089,
|
| 253 |
+
"eval_loss": 8.7124662399292,
|
| 254 |
+
"eval_runtime": 49.9244,
|
| 255 |
+
"eval_samples_per_second": 378.893,
|
| 256 |
+
"eval_steps_per_second": 1.482,
|
| 257 |
+
"step": 7500
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.2804950738052663,
|
| 261 |
+
"grad_norm": 3.34848690032959,
|
| 262 |
+
"learning_rate": 1.4018232819074336e-05,
|
| 263 |
+
"loss": 9.6994,
|
| 264 |
+
"step": 8000
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.2804950738052663,
|
| 268 |
+
"eval_cosine_accuracy": 0.9511532187461853,
|
| 269 |
+
"eval_loss": 8.740606307983398,
|
| 270 |
+
"eval_runtime": 51.2616,
|
| 271 |
+
"eval_samples_per_second": 369.009,
|
| 272 |
+
"eval_steps_per_second": 1.444,
|
| 273 |
+
"step": 8000
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.29802601591809547,
|
| 277 |
+
"grad_norm": 3.3112778663635254,
|
| 278 |
+
"learning_rate": 1.4894810659186536e-05,
|
| 279 |
+
"loss": 9.6911,
|
| 280 |
+
"step": 8500
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.29802601591809547,
|
| 284 |
+
"eval_cosine_accuracy": 0.9524672627449036,
|
| 285 |
+
"eval_loss": 8.709793090820312,
|
| 286 |
+
"eval_runtime": 50.3388,
|
| 287 |
+
"eval_samples_per_second": 375.774,
|
| 288 |
+
"eval_steps_per_second": 1.47,
|
| 289 |
+
"step": 8500
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.3155569580309246,
|
| 293 |
+
"grad_norm": 3.034113883972168,
|
| 294 |
+
"learning_rate": 1.577138849929874e-05,
|
| 295 |
+
"loss": 9.6785,
|
| 296 |
+
"step": 9000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.3155569580309246,
|
| 300 |
+
"eval_cosine_accuracy": 0.9522860050201416,
|
| 301 |
+
"eval_loss": 8.736331939697266,
|
| 302 |
+
"eval_runtime": 49.8153,
|
| 303 |
+
"eval_samples_per_second": 379.722,
|
| 304 |
+
"eval_steps_per_second": 1.485,
|
| 305 |
+
"step": 9000
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.3330879001437537,
|
| 309 |
+
"grad_norm": 3.5567679405212402,
|
| 310 |
+
"learning_rate": 1.664621318373072e-05,
|
| 311 |
+
"loss": 9.6627,
|
| 312 |
+
"step": 9500
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 0.3330879001437537,
|
| 316 |
+
"eval_cosine_accuracy": 0.9511532187461853,
|
| 317 |
+
"eval_loss": 8.755115509033203,
|
| 318 |
+
"eval_runtime": 50.7728,
|
| 319 |
+
"eval_samples_per_second": 372.562,
|
| 320 |
+
"eval_steps_per_second": 1.457,
|
| 321 |
+
"step": 9500
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.35061884225658285,
|
| 325 |
+
"grad_norm": 3.8073484897613525,
|
| 326 |
+
"learning_rate": 1.7522791023842917e-05,
|
| 327 |
+
"loss": 9.6493,
|
| 328 |
+
"step": 10000
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.35061884225658285,
|
| 332 |
+
"eval_cosine_accuracy": 0.9495219588279724,
|
| 333 |
+
"eval_loss": 8.82080078125,
|
| 334 |
+
"eval_runtime": 49.6151,
|
| 335 |
+
"eval_samples_per_second": 381.255,
|
| 336 |
+
"eval_steps_per_second": 1.491,
|
| 337 |
+
"step": 10000
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 0.36814978436941204,
|
| 341 |
+
"grad_norm": 3.8539516925811768,
|
| 342 |
+
"learning_rate": 1.8399368863955122e-05,
|
| 343 |
+
"loss": 9.6422,
|
| 344 |
+
"step": 10500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.36814978436941204,
|
| 348 |
+
"eval_cosine_accuracy": 0.9494766592979431,
|
| 349 |
+
"eval_loss": 8.80677318572998,
|
| 350 |
+
"eval_runtime": 51.2555,
|
| 351 |
+
"eval_samples_per_second": 369.053,
|
| 352 |
+
"eval_steps_per_second": 1.444,
|
| 353 |
+
"step": 10500
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.38568072648224117,
|
| 357 |
+
"grad_norm": 3.997718334197998,
|
| 358 |
+
"learning_rate": 1.9275946704067323e-05,
|
| 359 |
+
"loss": 9.6288,
|
| 360 |
+
"step": 11000
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.38568072648224117,
|
| 364 |
+
"eval_cosine_accuracy": 0.948706328868866,
|
| 365 |
+
"eval_loss": 8.862730026245117,
|
| 366 |
+
"eval_runtime": 50.5188,
|
| 367 |
+
"eval_samples_per_second": 374.435,
|
| 368 |
+
"eval_steps_per_second": 1.465,
|
| 369 |
+
"step": 11000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.4032116685950703,
|
| 373 |
+
"grad_norm": 4.089541435241699,
|
| 374 |
+
"learning_rate": 1.9999964570084554e-05,
|
| 375 |
+
"loss": 9.6166,
|
| 376 |
+
"step": 11500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 0.4032116685950703,
|
| 380 |
+
"eval_cosine_accuracy": 0.9495672583580017,
|
| 381 |
+
"eval_loss": 8.826420783996582,
|
| 382 |
+
"eval_runtime": 49.7378,
|
| 383 |
+
"eval_samples_per_second": 380.314,
|
| 384 |
+
"eval_steps_per_second": 1.488,
|
| 385 |
+
"step": 11500
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 0.4207426107078994,
|
| 389 |
+
"grad_norm": 3.3106064796447754,
|
| 390 |
+
"learning_rate": 1.9998387141025913e-05,
|
| 391 |
+
"loss": 9.6089,
|
| 392 |
+
"step": 12000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.4207426107078994,
|
| 396 |
+
"eval_cosine_accuracy": 0.9481625556945801,
|
| 397 |
+
"eval_loss": 8.905811309814453,
|
| 398 |
+
"eval_runtime": 50.5979,
|
| 399 |
+
"eval_samples_per_second": 373.849,
|
| 400 |
+
"eval_steps_per_second": 1.463,
|
| 401 |
+
"step": 12000
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.4382735528207286,
|
| 405 |
+
"grad_norm": 3.426849603652954,
|
| 406 |
+
"learning_rate": 1.99944696715976e-05,
|
| 407 |
+
"loss": 9.5992,
|
| 408 |
+
"step": 12500
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.4382735528207286,
|
| 412 |
+
"eval_cosine_accuracy": 0.9498844742774963,
|
| 413 |
+
"eval_loss": 8.883400917053223,
|
| 414 |
+
"eval_runtime": 50.1249,
|
| 415 |
+
"eval_samples_per_second": 377.377,
|
| 416 |
+
"eval_steps_per_second": 1.476,
|
| 417 |
+
"step": 12500
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.45580449493355774,
|
| 421 |
+
"grad_norm": 2.942359209060669,
|
| 422 |
+
"learning_rate": 1.9988213078651134e-05,
|
| 423 |
+
"loss": 9.5968,
|
| 424 |
+
"step": 13000
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 0.45580449493355774,
|
| 428 |
+
"eval_cosine_accuracy": 0.9498391151428223,
|
| 429 |
+
"eval_loss": 8.826370239257812,
|
| 430 |
+
"eval_runtime": 50.4882,
|
| 431 |
+
"eval_samples_per_second": 374.662,
|
| 432 |
+
"eval_steps_per_second": 1.466,
|
| 433 |
+
"step": 13000
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.47333543704638686,
|
| 437 |
+
"grad_norm": 3.6883912086486816,
|
| 438 |
+
"learning_rate": 1.997961882649071e-05,
|
| 439 |
+
"loss": 9.5859,
|
| 440 |
+
"step": 13500
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.47333543704638686,
|
| 444 |
+
"eval_cosine_accuracy": 0.947437584400177,
|
| 445 |
+
"eval_loss": 8.92893123626709,
|
| 446 |
+
"eval_runtime": 49.8701,
|
| 447 |
+
"eval_samples_per_second": 379.305,
|
| 448 |
+
"eval_steps_per_second": 1.484,
|
| 449 |
+
"step": 13500
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.490866379159216,
|
| 453 |
+
"grad_norm": 3.4566166400909424,
|
| 454 |
+
"learning_rate": 1.996871311569756e-05,
|
| 455 |
+
"loss": 9.5771,
|
| 456 |
+
"step": 14000
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 0.490866379159216,
|
| 460 |
+
"eval_cosine_accuracy": 0.9474828839302063,
|
| 461 |
+
"eval_loss": 8.924897193908691,
|
| 462 |
+
"eval_runtime": 49.7604,
|
| 463 |
+
"eval_samples_per_second": 380.142,
|
| 464 |
+
"eval_steps_per_second": 1.487,
|
| 465 |
+
"step": 14000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.5083973212720452,
|
| 469 |
+
"grad_norm": 4.192228317260742,
|
| 470 |
+
"learning_rate": 1.9955454789254234e-05,
|
| 471 |
+
"loss": 9.5657,
|
| 472 |
+
"step": 14500
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.5083973212720452,
|
| 476 |
+
"eval_cosine_accuracy": 0.9468938112258911,
|
| 477 |
+
"eval_loss": 8.962514877319336,
|
| 478 |
+
"eval_runtime": 51.4762,
|
| 479 |
+
"eval_samples_per_second": 367.47,
|
| 480 |
+
"eval_steps_per_second": 1.438,
|
| 481 |
+
"step": 14500
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 0.5259282633848743,
|
| 485 |
+
"grad_norm": 4.109299659729004,
|
| 486 |
+
"learning_rate": 1.9939866470405452e-05,
|
| 487 |
+
"loss": 9.5571,
|
| 488 |
+
"step": 15000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.5259282633848743,
|
| 492 |
+
"eval_cosine_accuracy": 0.9458969831466675,
|
| 493 |
+
"eval_loss": 8.966425895690918,
|
| 494 |
+
"eval_runtime": 50.6672,
|
| 495 |
+
"eval_samples_per_second": 373.338,
|
| 496 |
+
"eval_steps_per_second": 1.461,
|
| 497 |
+
"step": 15000
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 0.5434592054977034,
|
| 501 |
+
"grad_norm": 3.5839483737945557,
|
| 502 |
+
"learning_rate": 1.9921951807469183e-05,
|
| 503 |
+
"loss": 9.5559,
|
| 504 |
+
"step": 15500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.5434592054977034,
|
| 508 |
+
"eval_cosine_accuracy": 0.9465766549110413,
|
| 509 |
+
"eval_loss": 8.933329582214355,
|
| 510 |
+
"eval_runtime": 50.2462,
|
| 511 |
+
"eval_samples_per_second": 376.466,
|
| 512 |
+
"eval_steps_per_second": 1.473,
|
| 513 |
+
"step": 15500
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.5609901476105326,
|
| 517 |
+
"grad_norm": 3.516906499862671,
|
| 518 |
+
"learning_rate": 1.9901800560110894e-05,
|
| 519 |
+
"loss": 9.5514,
|
| 520 |
+
"step": 16000
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.5609901476105326,
|
| 524 |
+
"eval_cosine_accuracy": 0.9477094411849976,
|
| 525 |
+
"eval_loss": 8.930047988891602,
|
| 526 |
+
"eval_runtime": 51.0181,
|
| 527 |
+
"eval_samples_per_second": 370.771,
|
| 528 |
+
"eval_steps_per_second": 1.45,
|
| 529 |
+
"step": 16000
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.5785210897233617,
|
| 533 |
+
"grad_norm": 3.535010576248169,
|
| 534 |
+
"learning_rate": 1.9879255590325465e-05,
|
| 535 |
+
"loss": 9.5429,
|
| 536 |
+
"step": 16500
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.5785210897233617,
|
| 540 |
+
"eval_cosine_accuracy": 0.9487969279289246,
|
| 541 |
+
"eval_loss": 8.990011215209961,
|
| 542 |
+
"eval_runtime": 50.1912,
|
| 543 |
+
"eval_samples_per_second": 376.879,
|
| 544 |
+
"eval_steps_per_second": 1.474,
|
| 545 |
+
"step": 16500
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 0.5960520318361909,
|
| 549 |
+
"grad_norm": 3.881448268890381,
|
| 550 |
+
"learning_rate": 1.9854398461931087e-05,
|
| 551 |
+
"loss": 9.5319,
|
| 552 |
+
"step": 17000
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 0.5960520318361909,
|
| 556 |
+
"eval_cosine_accuracy": 0.943993866443634,
|
| 557 |
+
"eval_loss": 8.999488830566406,
|
| 558 |
+
"eval_runtime": 50.1438,
|
| 559 |
+
"eval_samples_per_second": 377.235,
|
| 560 |
+
"eval_steps_per_second": 1.476,
|
| 561 |
+
"step": 17000
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 0.6135829739490201,
|
| 565 |
+
"grad_norm": 3.682716131210327,
|
| 566 |
+
"learning_rate": 1.982723499253445e-05,
|
| 567 |
+
"loss": 9.5303,
|
| 568 |
+
"step": 17500
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 0.6135829739490201,
|
| 572 |
+
"eval_cosine_accuracy": 0.9459422826766968,
|
| 573 |
+
"eval_loss": 8.963994026184082,
|
| 574 |
+
"eval_runtime": 51.271,
|
| 575 |
+
"eval_samples_per_second": 368.941,
|
| 576 |
+
"eval_steps_per_second": 1.443,
|
| 577 |
+
"step": 17500
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.6311139160618492,
|
| 581 |
+
"grad_norm": 3.4688053131103516,
|
| 582 |
+
"learning_rate": 1.9797771539522402e-05,
|
| 583 |
+
"loss": 9.5292,
|
| 584 |
+
"step": 18000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.6311139160618492,
|
| 588 |
+
"eval_cosine_accuracy": 0.9459422826766968,
|
| 589 |
+
"eval_loss": 8.978798866271973,
|
| 590 |
+
"eval_runtime": 51.1172,
|
| 591 |
+
"eval_samples_per_second": 370.051,
|
| 592 |
+
"eval_steps_per_second": 1.448,
|
| 593 |
+
"step": 18000
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 0.6486448581746783,
|
| 597 |
+
"grad_norm": 4.622004985809326,
|
| 598 |
+
"learning_rate": 1.9766014998574075e-05,
|
| 599 |
+
"loss": 9.5179,
|
| 600 |
+
"step": 18500
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 0.6486448581746783,
|
| 604 |
+
"eval_cosine_accuracy": 0.9446282386779785,
|
| 605 |
+
"eval_loss": 9.016169548034668,
|
| 606 |
+
"eval_runtime": 50.2418,
|
| 607 |
+
"eval_samples_per_second": 376.499,
|
| 608 |
+
"eval_steps_per_second": 1.473,
|
| 609 |
+
"step": 18500
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.6661758002875074,
|
| 613 |
+
"grad_norm": 3.7142796516418457,
|
| 614 |
+
"learning_rate": 1.973197280204698e-05,
|
| 615 |
+
"loss": 9.5171,
|
| 616 |
+
"step": 19000
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 0.6661758002875074,
|
| 620 |
+
"eval_cosine_accuracy": 0.9446282386779785,
|
| 621 |
+
"eval_loss": 9.053177833557129,
|
| 622 |
+
"eval_runtime": 51.5419,
|
| 623 |
+
"eval_samples_per_second": 367.002,
|
| 624 |
+
"eval_steps_per_second": 1.436,
|
| 625 |
+
"step": 19000
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.6837067424003366,
|
| 629 |
+
"grad_norm": 3.596008539199829,
|
| 630 |
+
"learning_rate": 1.969565291723755e-05,
|
| 631 |
+
"loss": 9.509,
|
| 632 |
+
"step": 19500
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.6837067424003366,
|
| 636 |
+
"eval_cosine_accuracy": 0.9444922804832458,
|
| 637 |
+
"eval_loss": 9.100041389465332,
|
| 638 |
+
"eval_runtime": 50.8331,
|
| 639 |
+
"eval_samples_per_second": 372.12,
|
| 640 |
+
"eval_steps_per_second": 1.456,
|
| 641 |
+
"step": 19500
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.7012376845131657,
|
| 645 |
+
"grad_norm": 3.7148385047912598,
|
| 646 |
+
"learning_rate": 1.965706384451643e-05,
|
| 647 |
+
"loss": 9.4974,
|
| 648 |
+
"step": 20000
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 0.7012376845131657,
|
| 652 |
+
"eval_cosine_accuracy": 0.9441297650337219,
|
| 653 |
+
"eval_loss": 9.080968856811523,
|
| 654 |
+
"eval_runtime": 51.7342,
|
| 655 |
+
"eval_samples_per_second": 365.638,
|
| 656 |
+
"eval_steps_per_second": 1.43,
|
| 657 |
+
"step": 20000
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 0.7187686266259948,
|
| 661 |
+
"grad_norm": 4.166687965393066,
|
| 662 |
+
"learning_rate": 1.9616298563166923e-05,
|
| 663 |
+
"loss": 9.4997,
|
| 664 |
+
"step": 20500
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 0.7187686266259948,
|
| 668 |
+
"eval_cosine_accuracy": 0.9448094367980957,
|
| 669 |
+
"eval_loss": 9.160306930541992,
|
| 670 |
+
"eval_runtime": 50.2386,
|
| 671 |
+
"eval_samples_per_second": 376.523,
|
| 672 |
+
"eval_steps_per_second": 1.473,
|
| 673 |
+
"step": 20500
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 0.7362995687388241,
|
| 677 |
+
"grad_norm": 3.890392541885376,
|
| 678 |
+
"learning_rate": 1.9573203229260095e-05,
|
| 679 |
+
"loss": 9.4949,
|
| 680 |
+
"step": 21000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.7362995687388241,
|
| 684 |
+
"eval_cosine_accuracy": 0.9433141350746155,
|
| 685 |
+
"eval_loss": 9.170744895935059,
|
| 686 |
+
"eval_runtime": 50.2779,
|
| 687 |
+
"eval_samples_per_second": 376.229,
|
| 688 |
+
"eval_steps_per_second": 1.472,
|
| 689 |
+
"step": 21000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.7538305108516532,
|
| 693 |
+
"grad_norm": 4.466862678527832,
|
| 694 |
+
"learning_rate": 1.9527867365785086e-05,
|
| 695 |
+
"loss": 9.4909,
|
| 696 |
+
"step": 21500
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.7538305108516532,
|
| 700 |
+
"eval_cosine_accuracy": 0.942679762840271,
|
| 701 |
+
"eval_loss": 9.167838096618652,
|
| 702 |
+
"eval_runtime": 51.5157,
|
| 703 |
+
"eval_samples_per_second": 367.189,
|
| 704 |
+
"eval_steps_per_second": 1.436,
|
| 705 |
+
"step": 21500
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 0.7713614529644823,
|
| 709 |
+
"grad_norm": 3.551948308944702,
|
| 710 |
+
"learning_rate": 1.9480301583228286e-05,
|
| 711 |
+
"loss": 9.4909,
|
| 712 |
+
"step": 22000
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 0.7713614529644823,
|
| 716 |
+
"eval_cosine_accuracy": 0.9448094367980957,
|
| 717 |
+
"eval_loss": 9.07791519165039,
|
| 718 |
+
"eval_runtime": 50.9951,
|
| 719 |
+
"eval_samples_per_second": 370.938,
|
| 720 |
+
"eval_steps_per_second": 1.451,
|
| 721 |
+
"step": 22000
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 0.7888923950773115,
|
| 725 |
+
"grad_norm": 4.436821460723877,
|
| 726 |
+
"learning_rate": 1.943061878979968e-05,
|
| 727 |
+
"loss": 9.4826,
|
| 728 |
+
"step": 22500
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 0.7888923950773115,
|
| 732 |
+
"eval_cosine_accuracy": 0.9433594346046448,
|
| 733 |
+
"eval_loss": 9.1586275100708,
|
| 734 |
+
"eval_runtime": 50.3731,
|
| 735 |
+
"eval_samples_per_second": 375.518,
|
| 736 |
+
"eval_steps_per_second": 1.469,
|
| 737 |
+
"step": 22500
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.8064233371901406,
|
| 741 |
+
"grad_norm": 4.032012939453125,
|
| 742 |
+
"learning_rate": 1.9378631487807903e-05,
|
| 743 |
+
"loss": 9.4739,
|
| 744 |
+
"step": 23000
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.8064233371901406,
|
| 748 |
+
"eval_cosine_accuracy": 0.9438126087188721,
|
| 749 |
+
"eval_loss": 9.108503341674805,
|
| 750 |
+
"eval_runtime": 51.2155,
|
| 751 |
+
"eval_samples_per_second": 369.341,
|
| 752 |
+
"eval_steps_per_second": 1.445,
|
| 753 |
+
"step": 23000
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.8239542793029697,
|
| 757 |
+
"grad_norm": 3.9402658939361572,
|
| 758 |
+
"learning_rate": 1.932444919416554e-05,
|
| 759 |
+
"loss": 9.4761,
|
| 760 |
+
"step": 23500
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 0.8239542793029697,
|
| 764 |
+
"eval_cosine_accuracy": 0.9426344633102417,
|
| 765 |
+
"eval_loss": 9.16297721862793,
|
| 766 |
+
"eval_runtime": 50.324,
|
| 767 |
+
"eval_samples_per_second": 375.884,
|
| 768 |
+
"eval_steps_per_second": 1.47,
|
| 769 |
+
"step": 23500
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 0.8414852214157988,
|
| 773 |
+
"grad_norm": 4.0512261390686035,
|
| 774 |
+
"learning_rate": 1.9268199488264804e-05,
|
| 775 |
+
"loss": 9.4678,
|
| 776 |
+
"step": 24000
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 0.8414852214157988,
|
| 780 |
+
"eval_cosine_accuracy": 0.9408219456672668,
|
| 781 |
+
"eval_loss": 9.143158912658691,
|
| 782 |
+
"eval_runtime": 50.9267,
|
| 783 |
+
"eval_samples_per_second": 371.436,
|
| 784 |
+
"eval_steps_per_second": 1.453,
|
| 785 |
+
"step": 24000
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 0.859016163528628,
|
| 789 |
+
"grad_norm": 4.608471393585205,
|
| 790 |
+
"learning_rate": 1.920967008957015e-05,
|
| 791 |
+
"loss": 9.4634,
|
| 792 |
+
"step": 24500
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 0.859016163528628,
|
| 796 |
+
"eval_cosine_accuracy": 0.9429516792297363,
|
| 797 |
+
"eval_loss": 9.19399642944336,
|
| 798 |
+
"eval_runtime": 50.1418,
|
| 799 |
+
"eval_samples_per_second": 377.25,
|
| 800 |
+
"eval_steps_per_second": 1.476,
|
| 801 |
+
"step": 24500
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.8765471056414572,
|
| 805 |
+
"grad_norm": 3.4046618938446045,
|
| 806 |
+
"learning_rate": 1.914898524325209e-05,
|
| 807 |
+
"loss": 9.4592,
|
| 808 |
+
"step": 25000
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.8765471056414572,
|
| 812 |
+
"eval_cosine_accuracy": 0.942498505115509,
|
| 813 |
+
"eval_loss": 9.171710014343262,
|
| 814 |
+
"eval_runtime": 50.5114,
|
| 815 |
+
"eval_samples_per_second": 374.49,
|
| 816 |
+
"eval_steps_per_second": 1.465,
|
| 817 |
+
"step": 25000
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 0.8940780477542863,
|
| 821 |
+
"grad_norm": 4.221813678741455,
|
| 822 |
+
"learning_rate": 1.908615915210035e-05,
|
| 823 |
+
"loss": 9.458,
|
| 824 |
+
"step": 25500
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 0.8940780477542863,
|
| 828 |
+
"eval_cosine_accuracy": 0.9407313466072083,
|
| 829 |
+
"eval_loss": 9.206527709960938,
|
| 830 |
+
"eval_runtime": 51.7106,
|
| 831 |
+
"eval_samples_per_second": 365.805,
|
| 832 |
+
"eval_steps_per_second": 1.431,
|
| 833 |
+
"step": 25500
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 0.9116089898671155,
|
| 837 |
+
"grad_norm": 4.586436748504639,
|
| 838 |
+
"learning_rate": 1.902120652004543e-05,
|
| 839 |
+
"loss": 9.449,
|
| 840 |
+
"step": 26000
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 0.9116089898671155,
|
| 844 |
+
"eval_cosine_accuracy": 0.940504789352417,
|
| 845 |
+
"eval_loss": 9.243682861328125,
|
| 846 |
+
"eval_runtime": 50.8955,
|
| 847 |
+
"eval_samples_per_second": 371.664,
|
| 848 |
+
"eval_steps_per_second": 1.454,
|
| 849 |
+
"step": 26000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.9291399319799446,
|
| 853 |
+
"grad_norm": 4.309023380279541,
|
| 854 |
+
"learning_rate": 1.895414254871725e-05,
|
| 855 |
+
"loss": 9.4469,
|
| 856 |
+
"step": 26500
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.9291399319799446,
|
| 860 |
+
"eval_cosine_accuracy": 0.9420906901359558,
|
| 861 |
+
"eval_loss": 9.203714370727539,
|
| 862 |
+
"eval_runtime": 49.9746,
|
| 863 |
+
"eval_samples_per_second": 378.512,
|
| 864 |
+
"eval_steps_per_second": 1.481,
|
| 865 |
+
"step": 26500
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 0.9466708740927737,
|
| 869 |
+
"grad_norm": 3.725597858428955,
|
| 870 |
+
"learning_rate": 1.8884982933887337e-05,
|
| 871 |
+
"loss": 9.4501,
|
| 872 |
+
"step": 27000
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 0.9466708740927737,
|
| 876 |
+
"eval_cosine_accuracy": 0.9427704215049744,
|
| 877 |
+
"eval_loss": 9.19096565246582,
|
| 878 |
+
"eval_runtime": 51.4072,
|
| 879 |
+
"eval_samples_per_second": 367.964,
|
| 880 |
+
"eval_steps_per_second": 1.439,
|
| 881 |
+
"step": 27000
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 0.9642018162056029,
|
| 885 |
+
"grad_norm": 4.729601860046387,
|
| 886 |
+
"learning_rate": 1.8813743861795376e-05,
|
| 887 |
+
"loss": 9.4492,
|
| 888 |
+
"step": 27500
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 0.9642018162056029,
|
| 892 |
+
"eval_cosine_accuracy": 0.9398251175880432,
|
| 893 |
+
"eval_loss": 9.26838493347168,
|
| 894 |
+
"eval_runtime": 50.8114,
|
| 895 |
+
"eval_samples_per_second": 372.279,
|
| 896 |
+
"eval_steps_per_second": 1.456,
|
| 897 |
+
"step": 27500
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 0.981732758318432,
|
| 901 |
+
"grad_norm": 3.041431188583374,
|
| 902 |
+
"learning_rate": 1.8740590656409715e-05,
|
| 903 |
+
"loss": 9.445,
|
| 904 |
+
"step": 28000
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 0.981732758318432,
|
| 908 |
+
"eval_cosine_accuracy": 0.9418188333511353,
|
| 909 |
+
"eval_loss": 9.160036087036133,
|
| 910 |
+
"eval_runtime": 50.8238,
|
| 911 |
+
"eval_samples_per_second": 372.188,
|
| 912 |
+
"eval_steps_per_second": 1.456,
|
| 913 |
+
"step": 28000
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.9992637004312612,
|
| 917 |
+
"grad_norm": 3.983847141265869,
|
| 918 |
+
"learning_rate": 1.8665247245147084e-05,
|
| 919 |
+
"loss": 9.4356,
|
| 920 |
+
"step": 28500
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.9992637004312612,
|
| 924 |
+
"eval_cosine_accuracy": 0.9409579038619995,
|
| 925 |
+
"eval_loss": 9.164313316345215,
|
| 926 |
+
"eval_runtime": 50.3081,
|
| 927 |
+
"eval_samples_per_second": 376.003,
|
| 928 |
+
"eval_steps_per_second": 1.471,
|
| 929 |
+
"step": 28500
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 1.0167946425440904,
|
| 933 |
+
"grad_norm": 4.640812873840332,
|
| 934 |
+
"learning_rate": 1.8588032558854093e-05,
|
| 935 |
+
"loss": 9.4184,
|
| 936 |
+
"step": 29000
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 1.0167946425440904,
|
| 940 |
+
"eval_cosine_accuracy": 0.9420906901359558,
|
| 941 |
+
"eval_loss": 9.223180770874023,
|
| 942 |
+
"eval_runtime": 51.2132,
|
| 943 |
+
"eval_samples_per_second": 369.358,
|
| 944 |
+
"eval_steps_per_second": 1.445,
|
| 945 |
+
"step": 29000
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 1.0343255846569195,
|
| 949 |
+
"grad_norm": 4.3706278800964355,
|
| 950 |
+
"learning_rate": 1.8508655197341246e-05,
|
| 951 |
+
"loss": 9.4144,
|
| 952 |
+
"step": 29500
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 1.0343255846569195,
|
| 956 |
+
"eval_cosine_accuracy": 0.9394626021385193,
|
| 957 |
+
"eval_loss": 9.316435813903809,
|
| 958 |
+
"eval_runtime": 50.3418,
|
| 959 |
+
"eval_samples_per_second": 375.752,
|
| 960 |
+
"eval_steps_per_second": 1.47,
|
| 961 |
+
"step": 29500
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 1.0518565267697486,
|
| 965 |
+
"grad_norm": 3.948193073272705,
|
| 966 |
+
"learning_rate": 1.842728645498141e-05,
|
| 967 |
+
"loss": 9.4131,
|
| 968 |
+
"step": 30000
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 1.0518565267697486,
|
| 972 |
+
"eval_cosine_accuracy": 0.9401422739028931,
|
| 973 |
+
"eval_loss": 9.309089660644531,
|
| 974 |
+
"eval_runtime": 51.3612,
|
| 975 |
+
"eval_samples_per_second": 368.294,
|
| 976 |
+
"eval_steps_per_second": 1.441,
|
| 977 |
+
"step": 30000
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 1.0693874688825777,
|
| 981 |
+
"grad_norm": 5.29146671295166,
|
| 982 |
+
"learning_rate": 1.834394537546025e-05,
|
| 983 |
+
"loss": 9.4118,
|
| 984 |
+
"step": 30500
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 1.0693874688825777,
|
| 988 |
+
"eval_cosine_accuracy": 0.9416828751564026,
|
| 989 |
+
"eval_loss": 9.266592025756836,
|
| 990 |
+
"eval_runtime": 50.3878,
|
| 991 |
+
"eval_samples_per_second": 375.408,
|
| 992 |
+
"eval_steps_per_second": 1.469,
|
| 993 |
+
"step": 30500
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 1.0869184109954069,
|
| 997 |
+
"grad_norm": 4.25240421295166,
|
| 998 |
+
"learning_rate": 1.8258651464072758e-05,
|
| 999 |
+
"loss": 9.4104,
|
| 1000 |
+
"step": 31000
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 1.0869184109954069,
|
| 1004 |
+
"eval_cosine_accuracy": 0.9401422739028931,
|
| 1005 |
+
"eval_loss": 9.272114753723145,
|
| 1006 |
+
"eval_runtime": 51.7763,
|
| 1007 |
+
"eval_samples_per_second": 365.341,
|
| 1008 |
+
"eval_steps_per_second": 1.429,
|
| 1009 |
+
"step": 31000
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 1.104449353108236,
|
| 1013 |
+
"grad_norm": 5.2350311279296875,
|
| 1014 |
+
"learning_rate": 1.817142468315816e-05,
|
| 1015 |
+
"loss": 9.4056,
|
| 1016 |
+
"step": 31500
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 1.104449353108236,
|
| 1020 |
+
"eval_cosine_accuracy": 0.9388735294342041,
|
| 1021 |
+
"eval_loss": 9.314387321472168,
|
| 1022 |
+
"eval_runtime": 50.6506,
|
| 1023 |
+
"eval_samples_per_second": 373.461,
|
| 1024 |
+
"eval_steps_per_second": 1.461,
|
| 1025 |
+
"step": 31500
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.1219802952210651,
|
| 1029 |
+
"grad_norm": 4.285799503326416,
|
| 1030 |
+
"learning_rate": 1.8082285447427933e-05,
|
| 1031 |
+
"loss": 9.4055,
|
| 1032 |
+
"step": 32000
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.1219802952210651,
|
| 1036 |
+
"eval_cosine_accuracy": 0.9386016726493835,
|
| 1037 |
+
"eval_loss": 9.282620429992676,
|
| 1038 |
+
"eval_runtime": 51.0124,
|
| 1039 |
+
"eval_samples_per_second": 370.812,
|
| 1040 |
+
"eval_steps_per_second": 1.451,
|
| 1041 |
+
"step": 32000
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 1.1395112373338943,
|
| 1045 |
+
"grad_norm": 4.84486722946167,
|
| 1046 |
+
"learning_rate": 1.7991438554566148e-05,
|
| 1047 |
+
"loss": 9.4042,
|
| 1048 |
+
"step": 32500
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 1.1395112373338943,
|
| 1052 |
+
"eval_cosine_accuracy": 0.9377407431602478,
|
| 1053 |
+
"eval_loss": 9.319416046142578,
|
| 1054 |
+
"eval_runtime": 50.2012,
|
| 1055 |
+
"eval_samples_per_second": 376.804,
|
| 1056 |
+
"eval_steps_per_second": 1.474,
|
| 1057 |
+
"step": 32500
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 1.1570421794467234,
|
| 1061 |
+
"grad_norm": 3.767106533050537,
|
| 1062 |
+
"learning_rate": 1.789854115785457e-05,
|
| 1063 |
+
"loss": 9.4035,
|
| 1064 |
+
"step": 33000
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 1.1570421794467234,
|
| 1068 |
+
"eval_cosine_accuracy": 0.9393266439437866,
|
| 1069 |
+
"eval_loss": 9.259844779968262,
|
| 1070 |
+
"eval_runtime": 50.4864,
|
| 1071 |
+
"eval_samples_per_second": 374.675,
|
| 1072 |
+
"eval_steps_per_second": 1.466,
|
| 1073 |
+
"step": 33000
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 1.1745731215595527,
|
| 1077 |
+
"grad_norm": 5.839101791381836,
|
| 1078 |
+
"learning_rate": 1.7803795172474618e-05,
|
| 1079 |
+
"loss": 9.3958,
|
| 1080 |
+
"step": 33500
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 1.1745731215595527,
|
| 1084 |
+
"eval_cosine_accuracy": 0.9382844567298889,
|
| 1085 |
+
"eval_loss": 9.356642723083496,
|
| 1086 |
+
"eval_runtime": 52.1548,
|
| 1087 |
+
"eval_samples_per_second": 362.69,
|
| 1088 |
+
"eval_steps_per_second": 1.419,
|
| 1089 |
+
"step": 33500
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 1.1921040636723816,
|
| 1093 |
+
"grad_norm": 4.048542022705078,
|
| 1094 |
+
"learning_rate": 1.7707222772945786e-05,
|
| 1095 |
+
"loss": 9.3962,
|
| 1096 |
+
"step": 34000
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 1.1921040636723816,
|
| 1100 |
+
"eval_cosine_accuracy": 0.9371063709259033,
|
| 1101 |
+
"eval_loss": 9.368973731994629,
|
| 1102 |
+
"eval_runtime": 51.3806,
|
| 1103 |
+
"eval_samples_per_second": 368.154,
|
| 1104 |
+
"eval_steps_per_second": 1.44,
|
| 1105 |
+
"step": 34000
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 1.209635005785211,
|
| 1109 |
+
"grad_norm": 4.300061225891113,
|
| 1110 |
+
"learning_rate": 1.7609045098639108e-05,
|
| 1111 |
+
"loss": 9.3973,
|
| 1112 |
+
"step": 34500
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 1.209635005785211,
|
| 1116 |
+
"eval_cosine_accuracy": 0.9370610117912292,
|
| 1117 |
+
"eval_loss": 9.36530876159668,
|
| 1118 |
+
"eval_runtime": 50.7281,
|
| 1119 |
+
"eval_samples_per_second": 372.89,
|
| 1120 |
+
"eval_steps_per_second": 1.459,
|
| 1121 |
+
"step": 34500
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 1.2271659478980401,
|
| 1125 |
+
"grad_norm": 5.508227348327637,
|
| 1126 |
+
"learning_rate": 1.750889163723017e-05,
|
| 1127 |
+
"loss": 9.395,
|
| 1128 |
+
"step": 35000
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 1.2271659478980401,
|
| 1132 |
+
"eval_cosine_accuracy": 0.9367891550064087,
|
| 1133 |
+
"eval_loss": 9.298746109008789,
|
| 1134 |
+
"eval_runtime": 51.1857,
|
| 1135 |
+
"eval_samples_per_second": 369.557,
|
| 1136 |
+
"eval_steps_per_second": 1.446,
|
| 1137 |
+
"step": 35000
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.2446968900108693,
|
| 1141 |
+
"grad_norm": 3.7528862953186035,
|
| 1142 |
+
"learning_rate": 1.7406980781422036e-05,
|
| 1143 |
+
"loss": 9.3872,
|
| 1144 |
+
"step": 35500
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.2446968900108693,
|
| 1148 |
+
"eval_cosine_accuracy": 0.9366078972816467,
|
| 1149 |
+
"eval_loss": 9.427058219909668,
|
| 1150 |
+
"eval_runtime": 50.8986,
|
| 1151 |
+
"eval_samples_per_second": 371.641,
|
| 1152 |
+
"eval_steps_per_second": 1.454,
|
| 1153 |
+
"step": 35500
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 1.2622278321236984,
|
| 1157 |
+
"grad_norm": 5.674192905426025,
|
| 1158 |
+
"learning_rate": 1.7303336382613265e-05,
|
| 1159 |
+
"loss": 9.3857,
|
| 1160 |
+
"step": 36000
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 1.2622278321236984,
|
| 1164 |
+
"eval_cosine_accuracy": 0.9400516748428345,
|
| 1165 |
+
"eval_loss": 9.360187530517578,
|
| 1166 |
+
"eval_runtime": 51.4823,
|
| 1167 |
+
"eval_samples_per_second": 367.427,
|
| 1168 |
+
"eval_steps_per_second": 1.437,
|
| 1169 |
+
"step": 36000
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 1.2797587742365275,
|
| 1173 |
+
"grad_norm": 4.7365241050720215,
|
| 1174 |
+
"learning_rate": 1.719798269792392e-05,
|
| 1175 |
+
"loss": 9.3904,
|
| 1176 |
+
"step": 36500
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 1.2797587742365275,
|
| 1180 |
+
"eval_cosine_accuracy": 0.9380578994750977,
|
| 1181 |
+
"eval_loss": 9.40904426574707,
|
| 1182 |
+
"eval_runtime": 50.6761,
|
| 1183 |
+
"eval_samples_per_second": 373.273,
|
| 1184 |
+
"eval_steps_per_second": 1.46,
|
| 1185 |
+
"step": 36500
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 1.2972897163493566,
|
| 1189 |
+
"grad_norm": 4.287518501281738,
|
| 1190 |
+
"learning_rate": 1.7090944384518385e-05,
|
| 1191 |
+
"loss": 9.3838,
|
| 1192 |
+
"step": 37000
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 1.2972897163493566,
|
| 1196 |
+
"eval_cosine_accuracy": 0.9371063709259033,
|
| 1197 |
+
"eval_loss": 9.43030071258545,
|
| 1198 |
+
"eval_runtime": 51.2054,
|
| 1199 |
+
"eval_samples_per_second": 369.414,
|
| 1200 |
+
"eval_steps_per_second": 1.445,
|
| 1201 |
+
"step": 37000
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 1.3148206584621858,
|
| 1205 |
+
"grad_norm": 5.627056121826172,
|
| 1206 |
+
"learning_rate": 1.698246552902906e-05,
|
| 1207 |
+
"loss": 9.375,
|
| 1208 |
+
"step": 37500
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 1.3148206584621858,
|
| 1212 |
+
"eval_cosine_accuracy": 0.9366078972816467,
|
| 1213 |
+
"eval_loss": 9.414908409118652,
|
| 1214 |
+
"eval_runtime": 50.5873,
|
| 1215 |
+
"eval_samples_per_second": 373.928,
|
| 1216 |
+
"eval_steps_per_second": 1.463,
|
| 1217 |
+
"step": 37500
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 1.332351600575015,
|
| 1221 |
+
"grad_norm": 3.8458166122436523,
|
| 1222 |
+
"learning_rate": 1.687213674354592e-05,
|
| 1223 |
+
"loss": 9.3814,
|
| 1224 |
+
"step": 38000
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 1.332351600575015,
|
| 1228 |
+
"eval_cosine_accuracy": 0.9362907409667969,
|
| 1229 |
+
"eval_loss": 9.302230834960938,
|
| 1230 |
+
"eval_runtime": 51.3911,
|
| 1231 |
+
"eval_samples_per_second": 368.079,
|
| 1232 |
+
"eval_steps_per_second": 1.44,
|
| 1233 |
+
"step": 38000
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 1.349882542687844,
|
| 1237 |
+
"grad_norm": 4.458573818206787,
|
| 1238 |
+
"learning_rate": 1.6760199590915256e-05,
|
| 1239 |
+
"loss": 9.38,
|
| 1240 |
+
"step": 38500
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 1.349882542687844,
|
| 1244 |
+
"eval_cosine_accuracy": 0.9334813356399536,
|
| 1245 |
+
"eval_loss": 9.464284896850586,
|
| 1246 |
+
"eval_runtime": 50.6868,
|
| 1247 |
+
"eval_samples_per_second": 373.194,
|
| 1248 |
+
"eval_steps_per_second": 1.46,
|
| 1249 |
+
"step": 38500
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.3674134848006732,
|
| 1253 |
+
"grad_norm": 5.627968788146973,
|
| 1254 |
+
"learning_rate": 1.6646680269108032e-05,
|
| 1255 |
+
"loss": 9.3712,
|
| 1256 |
+
"step": 39000
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.3674134848006732,
|
| 1260 |
+
"eval_cosine_accuracy": 0.9354298114776611,
|
| 1261 |
+
"eval_loss": 9.414664268493652,
|
| 1262 |
+
"eval_runtime": 51.2376,
|
| 1263 |
+
"eval_samples_per_second": 369.182,
|
| 1264 |
+
"eval_steps_per_second": 1.444,
|
| 1265 |
+
"step": 39000
|
| 1266 |
+
},
|
| 1267 |
+
{
|
| 1268 |
+
"epoch": 1.3849444269135023,
|
| 1269 |
+
"grad_norm": 4.634047508239746,
|
| 1270 |
+
"learning_rate": 1.6531837030882992e-05,
|
| 1271 |
+
"loss": 9.3687,
|
| 1272 |
+
"step": 39500
|
| 1273 |
+
},
|
| 1274 |
+
{
|
| 1275 |
+
"epoch": 1.3849444269135023,
|
| 1276 |
+
"eval_cosine_accuracy": 0.9389641284942627,
|
| 1277 |
+
"eval_loss": 9.37267017364502,
|
| 1278 |
+
"eval_runtime": 50.5135,
|
| 1279 |
+
"eval_samples_per_second": 374.474,
|
| 1280 |
+
"eval_steps_per_second": 1.465,
|
| 1281 |
+
"step": 39500
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 1.4024753690263314,
|
| 1285 |
+
"grad_norm": 5.698931694030762,
|
| 1286 |
+
"learning_rate": 1.6415236469812062e-05,
|
| 1287 |
+
"loss": 9.3721,
|
| 1288 |
+
"step": 40000
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"epoch": 1.4024753690263314,
|
| 1292 |
+
"eval_cosine_accuracy": 0.9327563643455505,
|
| 1293 |
+
"eval_loss": 9.466745376586914,
|
| 1294 |
+
"eval_runtime": 51.109,
|
| 1295 |
+
"eval_samples_per_second": 370.111,
|
| 1296 |
+
"eval_steps_per_second": 1.448,
|
| 1297 |
+
"step": 40000
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 1.4200063111391605,
|
| 1301 |
+
"grad_norm": 5.904776096343994,
|
| 1302 |
+
"learning_rate": 1.6297134475349062e-05,
|
| 1303 |
+
"loss": 9.3665,
|
| 1304 |
+
"step": 40500
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 1.4200063111391605,
|
| 1308 |
+
"eval_cosine_accuracy": 0.936834454536438,
|
| 1309 |
+
"eval_loss": 9.418221473693848,
|
| 1310 |
+
"eval_runtime": 50.641,
|
| 1311 |
+
"eval_samples_per_second": 373.531,
|
| 1312 |
+
"eval_steps_per_second": 1.461,
|
| 1313 |
+
"step": 40500
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 1.4375372532519899,
|
| 1317 |
+
"grad_norm": 4.435075283050537,
|
| 1318 |
+
"learning_rate": 1.617755868829554e-05,
|
| 1319 |
+
"loss": 9.3679,
|
| 1320 |
+
"step": 41000
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 1.4375372532519899,
|
| 1324 |
+
"eval_cosine_accuracy": 0.9358376264572144,
|
| 1325 |
+
"eval_loss": 9.410283088684082,
|
| 1326 |
+
"eval_runtime": 51.0045,
|
| 1327 |
+
"eval_samples_per_second": 370.869,
|
| 1328 |
+
"eval_steps_per_second": 1.451,
|
| 1329 |
+
"step": 41000
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 1.4550681953648188,
|
| 1333 |
+
"grad_norm": 3.6580729484558105,
|
| 1334 |
+
"learning_rate": 1.60565370943821e-05,
|
| 1335 |
+
"loss": 9.367,
|
| 1336 |
+
"step": 41500
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 1.4550681953648188,
|
| 1340 |
+
"eval_cosine_accuracy": 0.934840738773346,
|
| 1341 |
+
"eval_loss": 9.38784408569336,
|
| 1342 |
+
"eval_runtime": 50.4355,
|
| 1343 |
+
"eval_samples_per_second": 375.053,
|
| 1344 |
+
"eval_steps_per_second": 1.467,
|
| 1345 |
+
"step": 41500
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 1.4725991374776481,
|
| 1349 |
+
"grad_norm": 5.021347522735596,
|
| 1350 |
+
"learning_rate": 1.593434429152759e-05,
|
| 1351 |
+
"loss": 9.3629,
|
| 1352 |
+
"step": 42000
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 1.4725991374776481,
|
| 1356 |
+
"eval_cosine_accuracy": 0.9335266947746277,
|
| 1357 |
+
"eval_loss": 9.44973373413086,
|
| 1358 |
+
"eval_runtime": 51.1238,
|
| 1359 |
+
"eval_samples_per_second": 370.003,
|
| 1360 |
+
"eval_steps_per_second": 1.447,
|
| 1361 |
+
"step": 42000
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.4901300795904773,
|
| 1365 |
+
"grad_norm": 4.628762722015381,
|
| 1366 |
+
"learning_rate": 1.5810519136812e-05,
|
| 1367 |
+
"loss": 9.3608,
|
| 1368 |
+
"step": 42500
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.4901300795904773,
|
| 1372 |
+
"eval_cosine_accuracy": 0.9351578950881958,
|
| 1373 |
+
"eval_loss": 9.441567420959473,
|
| 1374 |
+
"eval_runtime": 51.2421,
|
| 1375 |
+
"eval_samples_per_second": 369.15,
|
| 1376 |
+
"eval_steps_per_second": 1.444,
|
| 1377 |
+
"step": 42500
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 1.5076610217033064,
|
| 1381 |
+
"grad_norm": 5.742382526397705,
|
| 1382 |
+
"learning_rate": 1.568533407782813e-05,
|
| 1383 |
+
"loss": 9.3566,
|
| 1384 |
+
"step": 43000
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 1.5076610217033064,
|
| 1388 |
+
"eval_cosine_accuracy": 0.9342516660690308,
|
| 1389 |
+
"eval_loss": 9.518505096435547,
|
| 1390 |
+
"eval_runtime": 51.2382,
|
| 1391 |
+
"eval_samples_per_second": 369.178,
|
| 1392 |
+
"eval_steps_per_second": 1.444,
|
| 1393 |
+
"step": 43000
|
| 1394 |
+
},
|
| 1395 |
+
{
|
| 1396 |
+
"epoch": 1.5251919638161355,
|
| 1397 |
+
"grad_norm": 3.848411798477173,
|
| 1398 |
+
"learning_rate": 1.5558818413110603e-05,
|
| 1399 |
+
"loss": 9.3529,
|
| 1400 |
+
"step": 43500
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 1.5251919638161355,
|
| 1404 |
+
"eval_cosine_accuracy": 0.9349766373634338,
|
| 1405 |
+
"eval_loss": 9.524738311767578,
|
| 1406 |
+
"eval_runtime": 50.4912,
|
| 1407 |
+
"eval_samples_per_second": 374.64,
|
| 1408 |
+
"eval_steps_per_second": 1.466,
|
| 1409 |
+
"step": 43500
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 1.5427229059289647,
|
| 1413 |
+
"grad_norm": 6.790402889251709,
|
| 1414 |
+
"learning_rate": 1.543100175261136e-05,
|
| 1415 |
+
"loss": 9.3562,
|
| 1416 |
+
"step": 44000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 1.5427229059289647,
|
| 1420 |
+
"eval_cosine_accuracy": 0.9335266947746277,
|
| 1421 |
+
"eval_loss": 9.521926879882812,
|
| 1422 |
+
"eval_runtime": 51.1456,
|
| 1423 |
+
"eval_samples_per_second": 369.846,
|
| 1424 |
+
"eval_steps_per_second": 1.447,
|
| 1425 |
+
"step": 44000
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 1.5602538480417938,
|
| 1429 |
+
"grad_norm": 6.224915504455566,
|
| 1430 |
+
"learning_rate": 1.5302173434760524e-05,
|
| 1431 |
+
"loss": 9.348,
|
| 1432 |
+
"step": 44500
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 1.5602538480417938,
|
| 1436 |
+
"eval_cosine_accuracy": 0.9317141771316528,
|
| 1437 |
+
"eval_loss": 9.596720695495605,
|
| 1438 |
+
"eval_runtime": 50.4506,
|
| 1439 |
+
"eval_samples_per_second": 374.941,
|
| 1440 |
+
"eval_steps_per_second": 1.467,
|
| 1441 |
+
"step": 44500
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 1.577784790154623,
|
| 1445 |
+
"grad_norm": 4.383378505706787,
|
| 1446 |
+
"learning_rate": 1.51718472748951e-05,
|
| 1447 |
+
"loss": 9.351,
|
| 1448 |
+
"step": 45000
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 1.577784790154623,
|
| 1452 |
+
"eval_cosine_accuracy": 0.9352032542228699,
|
| 1453 |
+
"eval_loss": 9.493390083312988,
|
| 1454 |
+
"eval_runtime": 51.2184,
|
| 1455 |
+
"eval_samples_per_second": 369.32,
|
| 1456 |
+
"eval_steps_per_second": 1.445,
|
| 1457 |
+
"step": 45000
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 1.595315732267452,
|
| 1461 |
+
"grad_norm": 3.945154905319214,
|
| 1462 |
+
"learning_rate": 1.5040310686663743e-05,
|
| 1463 |
+
"loss": 9.3453,
|
| 1464 |
+
"step": 45500
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 1.595315732267452,
|
| 1468 |
+
"eval_cosine_accuracy": 0.9343876242637634,
|
| 1469 |
+
"eval_loss": 9.535852432250977,
|
| 1470 |
+
"eval_runtime": 50.2244,
|
| 1471 |
+
"eval_samples_per_second": 376.63,
|
| 1472 |
+
"eval_steps_per_second": 1.473,
|
| 1473 |
+
"step": 45500
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.6128466743802812,
|
| 1477 |
+
"grad_norm": 5.017545223236084,
|
| 1478 |
+
"learning_rate": 1.4907594455124307e-05,
|
| 1479 |
+
"loss": 9.3495,
|
| 1480 |
+
"step": 46000
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.6128466743802812,
|
| 1484 |
+
"eval_cosine_accuracy": 0.9350672960281372,
|
| 1485 |
+
"eval_loss": 9.474373817443848,
|
| 1486 |
+
"eval_runtime": 50.8725,
|
| 1487 |
+
"eval_samples_per_second": 371.831,
|
| 1488 |
+
"eval_steps_per_second": 1.455,
|
| 1489 |
+
"step": 46000
|
| 1490 |
+
},
|
| 1491 |
+
{
|
| 1492 |
+
"epoch": 1.6303776164931103,
|
| 1493 |
+
"grad_norm": 4.84283447265625,
|
| 1494 |
+
"learning_rate": 1.4773729641420465e-05,
|
| 1495 |
+
"loss": 9.3431,
|
| 1496 |
+
"step": 46500
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 1.6303776164931103,
|
| 1500 |
+
"eval_cosine_accuracy": 0.9334360361099243,
|
| 1501 |
+
"eval_loss": 9.547622680664062,
|
| 1502 |
+
"eval_runtime": 50.1008,
|
| 1503 |
+
"eval_samples_per_second": 377.559,
|
| 1504 |
+
"eval_steps_per_second": 1.477,
|
| 1505 |
+
"step": 46500
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 1.6479085586059394,
|
| 1509 |
+
"grad_norm": 3.5967538356781006,
|
| 1510 |
+
"learning_rate": 1.4639018633706608e-05,
|
| 1511 |
+
"loss": 9.3469,
|
| 1512 |
+
"step": 47000
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 1.6479085586059394,
|
| 1516 |
+
"eval_cosine_accuracy": 0.9349313378334045,
|
| 1517 |
+
"eval_loss": 9.48145866394043,
|
| 1518 |
+
"eval_runtime": 50.9794,
|
| 1519 |
+
"eval_samples_per_second": 371.052,
|
| 1520 |
+
"eval_steps_per_second": 1.452,
|
| 1521 |
+
"step": 47000
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 1.6654395007187688,
|
| 1525 |
+
"grad_norm": 5.372916221618652,
|
| 1526 |
+
"learning_rate": 1.450295304666048e-05,
|
| 1527 |
+
"loss": 9.344,
|
| 1528 |
+
"step": 47500
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 1.6654395007187688,
|
| 1532 |
+
"eval_cosine_accuracy": 0.9335266947746277,
|
| 1533 |
+
"eval_loss": 9.521085739135742,
|
| 1534 |
+
"eval_runtime": 50.4516,
|
| 1535 |
+
"eval_samples_per_second": 374.934,
|
| 1536 |
+
"eval_steps_per_second": 1.467,
|
| 1537 |
+
"step": 47500
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 1.6829704428315977,
|
| 1541 |
+
"grad_norm": 4.620992183685303,
|
| 1542 |
+
"learning_rate": 1.4365833580447305e-05,
|
| 1543 |
+
"loss": 9.3379,
|
| 1544 |
+
"step": 48000
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 1.6829704428315977,
|
| 1548 |
+
"eval_cosine_accuracy": 0.934840738773346,
|
| 1549 |
+
"eval_loss": 9.54494857788086,
|
| 1550 |
+
"eval_runtime": 50.8349,
|
| 1551 |
+
"eval_samples_per_second": 372.107,
|
| 1552 |
+
"eval_steps_per_second": 1.456,
|
| 1553 |
+
"step": 48000
|
| 1554 |
+
},
|
| 1555 |
+
{
|
| 1556 |
+
"epoch": 1.700501384944427,
|
| 1557 |
+
"grad_norm": 4.837863922119141,
|
| 1558 |
+
"learning_rate": 1.422769232675167e-05,
|
| 1559 |
+
"loss": 9.3387,
|
| 1560 |
+
"step": 48500
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 1.700501384944427,
|
| 1564 |
+
"eval_cosine_accuracy": 0.935928225517273,
|
| 1565 |
+
"eval_loss": 9.489590644836426,
|
| 1566 |
+
"eval_runtime": 50.4729,
|
| 1567 |
+
"eval_samples_per_second": 374.775,
|
| 1568 |
+
"eval_steps_per_second": 1.466,
|
| 1569 |
+
"step": 48500
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 1.718032327057256,
|
| 1573 |
+
"grad_norm": 3.820568323135376,
|
| 1574 |
+
"learning_rate": 1.408856161639911e-05,
|
| 1575 |
+
"loss": 9.3345,
|
| 1576 |
+
"step": 49000
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 1.718032327057256,
|
| 1580 |
+
"eval_cosine_accuracy": 0.933753252029419,
|
| 1581 |
+
"eval_loss": 9.51984977722168,
|
| 1582 |
+
"eval_runtime": 51.0265,
|
| 1583 |
+
"eval_samples_per_second": 370.709,
|
| 1584 |
+
"eval_steps_per_second": 1.45,
|
| 1585 |
+
"step": 49000
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.7355632691700853,
|
| 1589 |
+
"grad_norm": 5.366570949554443,
|
| 1590 |
+
"learning_rate": 1.3948755120223541e-05,
|
| 1591 |
+
"loss": 9.3392,
|
| 1592 |
+
"step": 49500
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.7355632691700853,
|
| 1596 |
+
"eval_cosine_accuracy": 0.9317141771316528,
|
| 1597 |
+
"eval_loss": 9.579914093017578,
|
| 1598 |
+
"eval_runtime": 50.1951,
|
| 1599 |
+
"eval_samples_per_second": 376.849,
|
| 1600 |
+
"eval_steps_per_second": 1.474,
|
| 1601 |
+
"step": 49500
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"epoch": 1.7530942112829142,
|
| 1605 |
+
"grad_norm": 4.166445732116699,
|
| 1606 |
+
"learning_rate": 1.3807745223059599e-05,
|
| 1607 |
+
"loss": 9.3285,
|
| 1608 |
+
"step": 50000
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 1.7530942112829142,
|
| 1612 |
+
"eval_cosine_accuracy": 0.9311251044273376,
|
| 1613 |
+
"eval_loss": 9.662346839904785,
|
| 1614 |
+
"eval_runtime": 50.7663,
|
| 1615 |
+
"eval_samples_per_second": 372.609,
|
| 1616 |
+
"eval_steps_per_second": 1.458,
|
| 1617 |
+
"step": 50000
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 1.7706251533957436,
|
| 1621 |
+
"grad_norm": 3.8759758472442627,
|
| 1622 |
+
"learning_rate": 1.3665844154408324e-05,
|
| 1623 |
+
"loss": 9.332,
|
| 1624 |
+
"step": 50500
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"epoch": 1.7706251533957436,
|
| 1628 |
+
"eval_cosine_accuracy": 0.933390736579895,
|
| 1629 |
+
"eval_loss": 9.592278480529785,
|
| 1630 |
+
"eval_runtime": 50.1417,
|
| 1631 |
+
"eval_samples_per_second": 377.251,
|
| 1632 |
+
"eval_steps_per_second": 1.476,
|
| 1633 |
+
"step": 50500
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 1.7881560955085727,
|
| 1637 |
+
"grad_norm": 6.290439128875732,
|
| 1638 |
+
"learning_rate": 1.3523085125049056e-05,
|
| 1639 |
+
"loss": 9.3293,
|
| 1640 |
+
"step": 51000
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 1.7881560955085727,
|
| 1644 |
+
"eval_cosine_accuracy": 0.932257890701294,
|
| 1645 |
+
"eval_loss": 9.621561050415039,
|
| 1646 |
+
"eval_runtime": 51.075,
|
| 1647 |
+
"eval_samples_per_second": 370.357,
|
| 1648 |
+
"eval_steps_per_second": 1.449,
|
| 1649 |
+
"step": 51000
|
| 1650 |
+
},
|
| 1651 |
+
{
|
| 1652 |
+
"epoch": 1.8056870376214018,
|
| 1653 |
+
"grad_norm": 4.655262470245361,
|
| 1654 |
+
"learning_rate": 1.3379789514312285e-05,
|
| 1655 |
+
"loss": 9.3262,
|
| 1656 |
+
"step": 51500
|
| 1657 |
+
},
|
| 1658 |
+
{
|
| 1659 |
+
"epoch": 1.8056870376214018,
|
| 1660 |
+
"eval_cosine_accuracy": 0.9350219964981079,
|
| 1661 |
+
"eval_loss": 9.6589937210083,
|
| 1662 |
+
"eval_runtime": 50.2239,
|
| 1663 |
+
"eval_samples_per_second": 376.634,
|
| 1664 |
+
"eval_steps_per_second": 1.473,
|
| 1665 |
+
"step": 51500
|
| 1666 |
+
},
|
| 1667 |
+
{
|
| 1668 |
+
"epoch": 1.823217979734231,
|
| 1669 |
+
"grad_norm": 3.824714422225952,
|
| 1670 |
+
"learning_rate": 1.3235416539477435e-05,
|
| 1671 |
+
"loss": 9.3275,
|
| 1672 |
+
"step": 52000
|
| 1673 |
+
},
|
| 1674 |
+
{
|
| 1675 |
+
"epoch": 1.823217979734231,
|
| 1676 |
+
"eval_cosine_accuracy": 0.9330735206604004,
|
| 1677 |
+
"eval_loss": 9.576985359191895,
|
| 1678 |
+
"eval_runtime": 51.6017,
|
| 1679 |
+
"eval_samples_per_second": 366.577,
|
| 1680 |
+
"eval_steps_per_second": 1.434,
|
| 1681 |
+
"step": 52000
|
| 1682 |
+
},
|
| 1683 |
+
{
|
| 1684 |
+
"epoch": 1.84074892184706,
|
| 1685 |
+
"grad_norm": 6.556931018829346,
|
| 1686 |
+
"learning_rate": 1.3090286341981737e-05,
|
| 1687 |
+
"loss": 9.3303,
|
| 1688 |
+
"step": 52500
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 1.84074892184706,
|
| 1692 |
+
"eval_cosine_accuracy": 0.931034505367279,
|
| 1693 |
+
"eval_loss": 9.682831764221191,
|
| 1694 |
+
"eval_runtime": 50.4378,
|
| 1695 |
+
"eval_samples_per_second": 375.036,
|
| 1696 |
+
"eval_steps_per_second": 1.467,
|
| 1697 |
+
"step": 52500
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 1.8582798639598892,
|
| 1701 |
+
"grad_norm": 4.34053897857666,
|
| 1702 |
+
"learning_rate": 1.2944432888355614e-05,
|
| 1703 |
+
"loss": 9.3303,
|
| 1704 |
+
"step": 53000
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 1.8582798639598892,
|
| 1708 |
+
"eval_cosine_accuracy": 0.9347047805786133,
|
| 1709 |
+
"eval_loss": 9.639060974121094,
|
| 1710 |
+
"eval_runtime": 51.2836,
|
| 1711 |
+
"eval_samples_per_second": 368.851,
|
| 1712 |
+
"eval_steps_per_second": 1.443,
|
| 1713 |
+
"step": 53000
|
| 1714 |
+
},
|
| 1715 |
+
{
|
| 1716 |
+
"epoch": 1.8758108060727183,
|
| 1717 |
+
"grad_norm": 3.9927992820739746,
|
| 1718 |
+
"learning_rate": 1.2797890314401645e-05,
|
| 1719 |
+
"loss": 9.3301,
|
| 1720 |
+
"step": 53500
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 1.8758108060727183,
|
| 1724 |
+
"eval_cosine_accuracy": 0.9314423203468323,
|
| 1725 |
+
"eval_loss": 9.678288459777832,
|
| 1726 |
+
"eval_runtime": 50.7061,
|
| 1727 |
+
"eval_samples_per_second": 373.052,
|
| 1728 |
+
"eval_steps_per_second": 1.459,
|
| 1729 |
+
"step": 53500
|
| 1730 |
+
},
|
| 1731 |
+
{
|
| 1732 |
+
"epoch": 1.8933417481855475,
|
| 1733 |
+
"grad_norm": 4.368465423583984,
|
| 1734 |
+
"learning_rate": 1.2650692917205358e-05,
|
| 1735 |
+
"loss": 9.3216,
|
| 1736 |
+
"step": 54000
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 1.8933417481855475,
|
| 1740 |
+
"eval_cosine_accuracy": 0.9313969612121582,
|
| 1741 |
+
"eval_loss": 9.642071723937988,
|
| 1742 |
+
"eval_runtime": 51.1254,
|
| 1743 |
+
"eval_samples_per_second": 369.993,
|
| 1744 |
+
"eval_steps_per_second": 1.447,
|
| 1745 |
+
"step": 54000
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 1.9108726902983766,
|
| 1749 |
+
"grad_norm": 5.101983547210693,
|
| 1750 |
+
"learning_rate": 1.2503171378809456e-05,
|
| 1751 |
+
"loss": 9.3219,
|
| 1752 |
+
"step": 54500
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 1.9108726902983766,
|
| 1756 |
+
"eval_cosine_accuracy": 0.9311704039573669,
|
| 1757 |
+
"eval_loss": 9.62327766418457,
|
| 1758 |
+
"eval_runtime": 50.236,
|
| 1759 |
+
"eval_samples_per_second": 376.542,
|
| 1760 |
+
"eval_steps_per_second": 1.473,
|
| 1761 |
+
"step": 54500
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 1.928403632411206,
|
| 1765 |
+
"grad_norm": 3.6307313442230225,
|
| 1766 |
+
"learning_rate": 1.2354768968282219e-05,
|
| 1767 |
+
"loss": 9.3197,
|
| 1768 |
+
"step": 55000
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"epoch": 1.928403632411206,
|
| 1772 |
+
"eval_cosine_accuracy": 0.9341157078742981,
|
| 1773 |
+
"eval_loss": 9.659159660339355,
|
| 1774 |
+
"eval_runtime": 51.3646,
|
| 1775 |
+
"eval_samples_per_second": 368.269,
|
| 1776 |
+
"eval_steps_per_second": 1.441,
|
| 1777 |
+
"step": 55000
|
| 1778 |
+
},
|
| 1779 |
+
{
|
| 1780 |
+
"epoch": 1.9459345745240348,
|
| 1781 |
+
"grad_norm": 3.93852162361145,
|
| 1782 |
+
"learning_rate": 1.2206113877334564e-05,
|
| 1783 |
+
"loss": 9.3179,
|
| 1784 |
+
"step": 55500
|
| 1785 |
+
},
|
| 1786 |
+
{
|
| 1787 |
+
"epoch": 1.9459345745240348,
|
| 1788 |
+
"eval_cosine_accuracy": 0.93248450756073,
|
| 1789 |
+
"eval_loss": 9.677461624145508,
|
| 1790 |
+
"eval_runtime": 51.2076,
|
| 1791 |
+
"eval_samples_per_second": 369.398,
|
| 1792 |
+
"eval_steps_per_second": 1.445,
|
| 1793 |
+
"step": 55500
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 1.9634655166368642,
|
| 1797 |
+
"grad_norm": 5.185349464416504,
|
| 1798 |
+
"learning_rate": 1.2056645097125428e-05,
|
| 1799 |
+
"loss": 9.3176,
|
| 1800 |
+
"step": 56000
|
| 1801 |
+
},
|
| 1802 |
+
{
|
| 1803 |
+
"epoch": 1.9634655166368642,
|
| 1804 |
+
"eval_cosine_accuracy": 0.9333454370498657,
|
| 1805 |
+
"eval_loss": 9.626531600952148,
|
| 1806 |
+
"eval_runtime": 50.8668,
|
| 1807 |
+
"eval_samples_per_second": 371.873,
|
| 1808 |
+
"eval_steps_per_second": 1.455,
|
| 1809 |
+
"step": 56000
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 1.980996458749693,
|
| 1813 |
+
"grad_norm": 5.199561595916748,
|
| 1814 |
+
"learning_rate": 1.1906694976027455e-05,
|
| 1815 |
+
"loss": 9.3139,
|
| 1816 |
+
"step": 56500
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 1.980996458749693,
|
| 1820 |
+
"eval_cosine_accuracy": 0.932121992111206,
|
| 1821 |
+
"eval_loss": 9.674995422363281,
|
| 1822 |
+
"eval_runtime": 50.0197,
|
| 1823 |
+
"eval_samples_per_second": 378.171,
|
| 1824 |
+
"eval_steps_per_second": 1.479,
|
| 1825 |
+
"step": 56500
|
| 1826 |
+
},
|
| 1827 |
+
{
|
| 1828 |
+
"epoch": 1.9985274008625225,
|
| 1829 |
+
"grad_norm": 4.581361770629883,
|
| 1830 |
+
"learning_rate": 1.1756298608634592e-05,
|
| 1831 |
+
"loss": 9.3101,
|
| 1832 |
+
"step": 57000
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"epoch": 1.9985274008625225,
|
| 1836 |
+
"eval_cosine_accuracy": 0.9318501353263855,
|
| 1837 |
+
"eval_loss": 9.694042205810547,
|
| 1838 |
+
"eval_runtime": 50.6442,
|
| 1839 |
+
"eval_samples_per_second": 373.508,
|
| 1840 |
+
"eval_steps_per_second": 1.461,
|
| 1841 |
+
"step": 57000
|
| 1842 |
+
}
|
| 1843 |
+
],
|
| 1844 |
+
"logging_steps": 500,
|
| 1845 |
+
"max_steps": 114084,
|
| 1846 |
+
"num_input_tokens_seen": 0,
|
| 1847 |
+
"num_train_epochs": 4,
|
| 1848 |
+
"save_steps": 500,
|
| 1849 |
+
"stateful_callbacks": {
|
| 1850 |
+
"TrainerControl": {
|
| 1851 |
+
"args": {
|
| 1852 |
+
"should_epoch_stop": false,
|
| 1853 |
+
"should_evaluate": false,
|
| 1854 |
+
"should_log": false,
|
| 1855 |
+
"should_save": true,
|
| 1856 |
+
"should_training_stop": false
|
| 1857 |
+
},
|
| 1858 |
+
"attributes": {}
|
| 1859 |
+
}
|
| 1860 |
+
},
|
| 1861 |
+
"total_flos": 0.0,
|
| 1862 |
+
"train_batch_size": 256,
|
| 1863 |
+
"trial_name": null,
|
| 1864 |
+
"trial_params": null
|
| 1865 |
+
}
|
checkpoint-57042/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ff251d0259f667a93594ea1efe197ff3b0827f9a1ffe30b0b91001733ac6ff94
|
| 3 |
+
size 5624
|
checkpoint-57042/training_metrics.json
ADDED
|
@@ -0,0 +1,690 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metrics_history": [
|
| 3 |
+
{
|
| 4 |
+
"epoch": 0.017530942112829143,
|
| 5 |
+
"global_step": 500,
|
| 6 |
+
"eval_loss": 10.808653831481934,
|
| 7 |
+
"eval_cosine_accuracy": 0.9304001331329346
|
| 8 |
+
},
|
| 9 |
+
{
|
| 10 |
+
"epoch": 0.035061884225658285,
|
| 11 |
+
"global_step": 1000,
|
| 12 |
+
"eval_loss": 9.035820007324219,
|
| 13 |
+
"eval_cosine_accuracy": 0.9219266772270203
|
| 14 |
+
},
|
| 15 |
+
{
|
| 16 |
+
"epoch": 0.05259282633848743,
|
| 17 |
+
"global_step": 1500,
|
| 18 |
+
"eval_loss": 8.834137916564941,
|
| 19 |
+
"eval_cosine_accuracy": 0.9366985559463501
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"epoch": 0.07012376845131657,
|
| 23 |
+
"global_step": 2000,
|
| 24 |
+
"eval_loss": 8.768850326538086,
|
| 25 |
+
"eval_cosine_accuracy": 0.9407313466072083
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.08765471056414571,
|
| 29 |
+
"global_step": 2500,
|
| 30 |
+
"eval_loss": 8.751864433288574,
|
| 31 |
+
"eval_cosine_accuracy": 0.9402782320976257
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.10518565267697486,
|
| 35 |
+
"global_step": 3000,
|
| 36 |
+
"eval_loss": 8.732973098754883,
|
| 37 |
+
"eval_cosine_accuracy": 0.9425438642501831
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.122716594789804,
|
| 41 |
+
"global_step": 3500,
|
| 42 |
+
"eval_loss": 8.691949844360352,
|
| 43 |
+
"eval_cosine_accuracy": 0.9468938112258911
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 0.14024753690263314,
|
| 47 |
+
"global_step": 4000,
|
| 48 |
+
"eval_loss": 8.692930221557617,
|
| 49 |
+
"eval_cosine_accuracy": 0.9477094411849976
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.1577784790154623,
|
| 53 |
+
"global_step": 4500,
|
| 54 |
+
"eval_loss": 8.666604995727539,
|
| 55 |
+
"eval_cosine_accuracy": 0.9502469301223755
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.17530942112829143,
|
| 59 |
+
"global_step": 5000,
|
| 60 |
+
"eval_loss": 8.687457084655762,
|
| 61 |
+
"eval_cosine_accuracy": 0.9490235447883606
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.19284036324112058,
|
| 65 |
+
"global_step": 5500,
|
| 66 |
+
"eval_loss": 8.682653427124023,
|
| 67 |
+
"eval_cosine_accuracy": 0.9515156745910645
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.2103713053539497,
|
| 71 |
+
"global_step": 6000,
|
| 72 |
+
"eval_loss": 8.673870086669922,
|
| 73 |
+
"eval_cosine_accuracy": 0.9527844190597534
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.22790224746677887,
|
| 77 |
+
"global_step": 6500,
|
| 78 |
+
"eval_loss": 8.690710067749023,
|
| 79 |
+
"eval_cosine_accuracy": 0.9525125622749329
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.245433189579608,
|
| 83 |
+
"global_step": 7000,
|
| 84 |
+
"eval_loss": 8.720589637756348,
|
| 85 |
+
"eval_cosine_accuracy": 0.9523313045501709
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.26296413169243715,
|
| 89 |
+
"global_step": 7500,
|
| 90 |
+
"eval_loss": 8.7124662399292,
|
| 91 |
+
"eval_cosine_accuracy": 0.9521500468254089
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.2804950738052663,
|
| 95 |
+
"global_step": 8000,
|
| 96 |
+
"eval_loss": 8.740606307983398,
|
| 97 |
+
"eval_cosine_accuracy": 0.9511532187461853
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.29802601591809547,
|
| 101 |
+
"global_step": 8500,
|
| 102 |
+
"eval_loss": 8.709793090820312,
|
| 103 |
+
"eval_cosine_accuracy": 0.9524672627449036
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 0.3155569580309246,
|
| 107 |
+
"global_step": 9000,
|
| 108 |
+
"eval_loss": 8.736331939697266,
|
| 109 |
+
"eval_cosine_accuracy": 0.9522860050201416
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.3330879001437537,
|
| 113 |
+
"global_step": 9500,
|
| 114 |
+
"eval_loss": 8.755115509033203,
|
| 115 |
+
"eval_cosine_accuracy": 0.9511532187461853
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.35061884225658285,
|
| 119 |
+
"global_step": 10000,
|
| 120 |
+
"eval_loss": 8.82080078125,
|
| 121 |
+
"eval_cosine_accuracy": 0.9495219588279724
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.36814978436941204,
|
| 125 |
+
"global_step": 10500,
|
| 126 |
+
"eval_loss": 8.80677318572998,
|
| 127 |
+
"eval_cosine_accuracy": 0.9494766592979431
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 0.38568072648224117,
|
| 131 |
+
"global_step": 11000,
|
| 132 |
+
"eval_loss": 8.862730026245117,
|
| 133 |
+
"eval_cosine_accuracy": 0.948706328868866
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 0.4032116685950703,
|
| 137 |
+
"global_step": 11500,
|
| 138 |
+
"eval_loss": 8.826420783996582,
|
| 139 |
+
"eval_cosine_accuracy": 0.9495672583580017
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.4207426107078994,
|
| 143 |
+
"global_step": 12000,
|
| 144 |
+
"eval_loss": 8.905811309814453,
|
| 145 |
+
"eval_cosine_accuracy": 0.9481625556945801
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.4382735528207286,
|
| 149 |
+
"global_step": 12500,
|
| 150 |
+
"eval_loss": 8.883400917053223,
|
| 151 |
+
"eval_cosine_accuracy": 0.9498844742774963
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.45580449493355774,
|
| 155 |
+
"global_step": 13000,
|
| 156 |
+
"eval_loss": 8.826370239257812,
|
| 157 |
+
"eval_cosine_accuracy": 0.9498391151428223
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.47333543704638686,
|
| 161 |
+
"global_step": 13500,
|
| 162 |
+
"eval_loss": 8.92893123626709,
|
| 163 |
+
"eval_cosine_accuracy": 0.947437584400177
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.490866379159216,
|
| 167 |
+
"global_step": 14000,
|
| 168 |
+
"eval_loss": 8.924897193908691,
|
| 169 |
+
"eval_cosine_accuracy": 0.9474828839302063
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.5083973212720452,
|
| 173 |
+
"global_step": 14500,
|
| 174 |
+
"eval_loss": 8.962514877319336,
|
| 175 |
+
"eval_cosine_accuracy": 0.9468938112258911
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 0.5259282633848743,
|
| 179 |
+
"global_step": 15000,
|
| 180 |
+
"eval_loss": 8.966425895690918,
|
| 181 |
+
"eval_cosine_accuracy": 0.9458969831466675
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.5434592054977034,
|
| 185 |
+
"global_step": 15500,
|
| 186 |
+
"eval_loss": 8.933329582214355,
|
| 187 |
+
"eval_cosine_accuracy": 0.9465766549110413
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 0.5609901476105326,
|
| 191 |
+
"global_step": 16000,
|
| 192 |
+
"eval_loss": 8.930047988891602,
|
| 193 |
+
"eval_cosine_accuracy": 0.9477094411849976
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.5785210897233617,
|
| 197 |
+
"global_step": 16500,
|
| 198 |
+
"eval_loss": 8.990011215209961,
|
| 199 |
+
"eval_cosine_accuracy": 0.9487969279289246
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.5960520318361909,
|
| 203 |
+
"global_step": 17000,
|
| 204 |
+
"eval_loss": 8.999488830566406,
|
| 205 |
+
"eval_cosine_accuracy": 0.943993866443634
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.6135829739490201,
|
| 209 |
+
"global_step": 17500,
|
| 210 |
+
"eval_loss": 8.963994026184082,
|
| 211 |
+
"eval_cosine_accuracy": 0.9459422826766968
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.6311139160618492,
|
| 215 |
+
"global_step": 18000,
|
| 216 |
+
"eval_loss": 8.978798866271973,
|
| 217 |
+
"eval_cosine_accuracy": 0.9459422826766968
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.6486448581746783,
|
| 221 |
+
"global_step": 18500,
|
| 222 |
+
"eval_loss": 9.016169548034668,
|
| 223 |
+
"eval_cosine_accuracy": 0.9446282386779785
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 0.6661758002875074,
|
| 227 |
+
"global_step": 19000,
|
| 228 |
+
"eval_loss": 9.053177833557129,
|
| 229 |
+
"eval_cosine_accuracy": 0.9446282386779785
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 0.6837067424003366,
|
| 233 |
+
"global_step": 19500,
|
| 234 |
+
"eval_loss": 9.100041389465332,
|
| 235 |
+
"eval_cosine_accuracy": 0.9444922804832458
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 0.7012376845131657,
|
| 239 |
+
"global_step": 20000,
|
| 240 |
+
"eval_loss": 9.080968856811523,
|
| 241 |
+
"eval_cosine_accuracy": 0.9441297650337219
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.7187686266259948,
|
| 245 |
+
"global_step": 20500,
|
| 246 |
+
"eval_loss": 9.160306930541992,
|
| 247 |
+
"eval_cosine_accuracy": 0.9448094367980957
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.7362995687388241,
|
| 251 |
+
"global_step": 21000,
|
| 252 |
+
"eval_loss": 9.170744895935059,
|
| 253 |
+
"eval_cosine_accuracy": 0.9433141350746155
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.7538305108516532,
|
| 257 |
+
"global_step": 21500,
|
| 258 |
+
"eval_loss": 9.167838096618652,
|
| 259 |
+
"eval_cosine_accuracy": 0.942679762840271
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 0.7713614529644823,
|
| 263 |
+
"global_step": 22000,
|
| 264 |
+
"eval_loss": 9.07791519165039,
|
| 265 |
+
"eval_cosine_accuracy": 0.9448094367980957
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.7888923950773115,
|
| 269 |
+
"global_step": 22500,
|
| 270 |
+
"eval_loss": 9.1586275100708,
|
| 271 |
+
"eval_cosine_accuracy": 0.9433594346046448
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.8064233371901406,
|
| 275 |
+
"global_step": 23000,
|
| 276 |
+
"eval_loss": 9.108503341674805,
|
| 277 |
+
"eval_cosine_accuracy": 0.9438126087188721
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 0.8239542793029697,
|
| 281 |
+
"global_step": 23500,
|
| 282 |
+
"eval_loss": 9.16297721862793,
|
| 283 |
+
"eval_cosine_accuracy": 0.9426344633102417
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.8414852214157988,
|
| 287 |
+
"global_step": 24000,
|
| 288 |
+
"eval_loss": 9.143158912658691,
|
| 289 |
+
"eval_cosine_accuracy": 0.9408219456672668
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.859016163528628,
|
| 293 |
+
"global_step": 24500,
|
| 294 |
+
"eval_loss": 9.19399642944336,
|
| 295 |
+
"eval_cosine_accuracy": 0.9429516792297363
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.8765471056414572,
|
| 299 |
+
"global_step": 25000,
|
| 300 |
+
"eval_loss": 9.171710014343262,
|
| 301 |
+
"eval_cosine_accuracy": 0.942498505115509
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 0.8940780477542863,
|
| 305 |
+
"global_step": 25500,
|
| 306 |
+
"eval_loss": 9.206527709960938,
|
| 307 |
+
"eval_cosine_accuracy": 0.9407313466072083
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 0.9116089898671155,
|
| 311 |
+
"global_step": 26000,
|
| 312 |
+
"eval_loss": 9.243682861328125,
|
| 313 |
+
"eval_cosine_accuracy": 0.940504789352417
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.9291399319799446,
|
| 317 |
+
"global_step": 26500,
|
| 318 |
+
"eval_loss": 9.203714370727539,
|
| 319 |
+
"eval_cosine_accuracy": 0.9420906901359558
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 0.9466708740927737,
|
| 323 |
+
"global_step": 27000,
|
| 324 |
+
"eval_loss": 9.19096565246582,
|
| 325 |
+
"eval_cosine_accuracy": 0.9427704215049744
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.9642018162056029,
|
| 329 |
+
"global_step": 27500,
|
| 330 |
+
"eval_loss": 9.26838493347168,
|
| 331 |
+
"eval_cosine_accuracy": 0.9398251175880432
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.981732758318432,
|
| 335 |
+
"global_step": 28000,
|
| 336 |
+
"eval_loss": 9.160036087036133,
|
| 337 |
+
"eval_cosine_accuracy": 0.9418188333511353
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 0.9992637004312612,
|
| 341 |
+
"global_step": 28500,
|
| 342 |
+
"eval_loss": 9.164313316345215,
|
| 343 |
+
"eval_cosine_accuracy": 0.9409579038619995
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 1.0167946425440904,
|
| 347 |
+
"global_step": 29000,
|
| 348 |
+
"eval_loss": 9.223180770874023,
|
| 349 |
+
"eval_cosine_accuracy": 0.9420906901359558
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 1.0343255846569195,
|
| 353 |
+
"global_step": 29500,
|
| 354 |
+
"eval_loss": 9.316435813903809,
|
| 355 |
+
"eval_cosine_accuracy": 0.9394626021385193
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 1.0518565267697486,
|
| 359 |
+
"global_step": 30000,
|
| 360 |
+
"eval_loss": 9.309089660644531,
|
| 361 |
+
"eval_cosine_accuracy": 0.9401422739028931
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 1.0693874688825777,
|
| 365 |
+
"global_step": 30500,
|
| 366 |
+
"eval_loss": 9.266592025756836,
|
| 367 |
+
"eval_cosine_accuracy": 0.9416828751564026
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.0869184109954069,
|
| 371 |
+
"global_step": 31000,
|
| 372 |
+
"eval_loss": 9.272114753723145,
|
| 373 |
+
"eval_cosine_accuracy": 0.9401422739028931
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 1.104449353108236,
|
| 377 |
+
"global_step": 31500,
|
| 378 |
+
"eval_loss": 9.314387321472168,
|
| 379 |
+
"eval_cosine_accuracy": 0.9388735294342041
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 1.1219802952210651,
|
| 383 |
+
"global_step": 32000,
|
| 384 |
+
"eval_loss": 9.282620429992676,
|
| 385 |
+
"eval_cosine_accuracy": 0.9386016726493835
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 1.1395112373338943,
|
| 389 |
+
"global_step": 32500,
|
| 390 |
+
"eval_loss": 9.319416046142578,
|
| 391 |
+
"eval_cosine_accuracy": 0.9377407431602478
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 1.1570421794467234,
|
| 395 |
+
"global_step": 33000,
|
| 396 |
+
"eval_loss": 9.259844779968262,
|
| 397 |
+
"eval_cosine_accuracy": 0.9393266439437866
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 1.1745731215595527,
|
| 401 |
+
"global_step": 33500,
|
| 402 |
+
"eval_loss": 9.356642723083496,
|
| 403 |
+
"eval_cosine_accuracy": 0.9382844567298889
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 1.1921040636723816,
|
| 407 |
+
"global_step": 34000,
|
| 408 |
+
"eval_loss": 9.368973731994629,
|
| 409 |
+
"eval_cosine_accuracy": 0.9371063709259033
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.209635005785211,
|
| 413 |
+
"global_step": 34500,
|
| 414 |
+
"eval_loss": 9.36530876159668,
|
| 415 |
+
"eval_cosine_accuracy": 0.9370610117912292
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.2271659478980401,
|
| 419 |
+
"global_step": 35000,
|
| 420 |
+
"eval_loss": 9.298746109008789,
|
| 421 |
+
"eval_cosine_accuracy": 0.9367891550064087
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 1.2446968900108693,
|
| 425 |
+
"global_step": 35500,
|
| 426 |
+
"eval_loss": 9.427058219909668,
|
| 427 |
+
"eval_cosine_accuracy": 0.9366078972816467
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 1.2622278321236984,
|
| 431 |
+
"global_step": 36000,
|
| 432 |
+
"eval_loss": 9.360187530517578,
|
| 433 |
+
"eval_cosine_accuracy": 0.9400516748428345
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 1.2797587742365275,
|
| 437 |
+
"global_step": 36500,
|
| 438 |
+
"eval_loss": 9.40904426574707,
|
| 439 |
+
"eval_cosine_accuracy": 0.9380578994750977
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 1.2972897163493566,
|
| 443 |
+
"global_step": 37000,
|
| 444 |
+
"eval_loss": 9.43030071258545,
|
| 445 |
+
"eval_cosine_accuracy": 0.9371063709259033
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 1.3148206584621858,
|
| 449 |
+
"global_step": 37500,
|
| 450 |
+
"eval_loss": 9.414908409118652,
|
| 451 |
+
"eval_cosine_accuracy": 0.9366078972816467
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.332351600575015,
|
| 455 |
+
"global_step": 38000,
|
| 456 |
+
"eval_loss": 9.302230834960938,
|
| 457 |
+
"eval_cosine_accuracy": 0.9362907409667969
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.349882542687844,
|
| 461 |
+
"global_step": 38500,
|
| 462 |
+
"eval_loss": 9.464284896850586,
|
| 463 |
+
"eval_cosine_accuracy": 0.9334813356399536
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 1.3674134848006732,
|
| 467 |
+
"global_step": 39000,
|
| 468 |
+
"eval_loss": 9.414664268493652,
|
| 469 |
+
"eval_cosine_accuracy": 0.9354298114776611
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 1.3849444269135023,
|
| 473 |
+
"global_step": 39500,
|
| 474 |
+
"eval_loss": 9.37267017364502,
|
| 475 |
+
"eval_cosine_accuracy": 0.9389641284942627
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 1.4024753690263314,
|
| 479 |
+
"global_step": 40000,
|
| 480 |
+
"eval_loss": 9.466745376586914,
|
| 481 |
+
"eval_cosine_accuracy": 0.9327563643455505
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 1.4200063111391605,
|
| 485 |
+
"global_step": 40500,
|
| 486 |
+
"eval_loss": 9.418221473693848,
|
| 487 |
+
"eval_cosine_accuracy": 0.936834454536438
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 1.4375372532519899,
|
| 491 |
+
"global_step": 41000,
|
| 492 |
+
"eval_loss": 9.410283088684082,
|
| 493 |
+
"eval_cosine_accuracy": 0.9358376264572144
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 1.4550681953648188,
|
| 497 |
+
"global_step": 41500,
|
| 498 |
+
"eval_loss": 9.38784408569336,
|
| 499 |
+
"eval_cosine_accuracy": 0.934840738773346
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.4725991374776481,
|
| 503 |
+
"global_step": 42000,
|
| 504 |
+
"eval_loss": 9.44973373413086,
|
| 505 |
+
"eval_cosine_accuracy": 0.9335266947746277
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 1.4901300795904773,
|
| 509 |
+
"global_step": 42500,
|
| 510 |
+
"eval_loss": 9.441567420959473,
|
| 511 |
+
"eval_cosine_accuracy": 0.9351578950881958
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 1.5076610217033064,
|
| 515 |
+
"global_step": 43000,
|
| 516 |
+
"eval_loss": 9.518505096435547,
|
| 517 |
+
"eval_cosine_accuracy": 0.9342516660690308
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 1.5251919638161355,
|
| 521 |
+
"global_step": 43500,
|
| 522 |
+
"eval_loss": 9.524738311767578,
|
| 523 |
+
"eval_cosine_accuracy": 0.9349766373634338
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 1.5427229059289647,
|
| 527 |
+
"global_step": 44000,
|
| 528 |
+
"eval_loss": 9.521926879882812,
|
| 529 |
+
"eval_cosine_accuracy": 0.9335266947746277
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 1.5602538480417938,
|
| 533 |
+
"global_step": 44500,
|
| 534 |
+
"eval_loss": 9.596720695495605,
|
| 535 |
+
"eval_cosine_accuracy": 0.9317141771316528
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 1.577784790154623,
|
| 539 |
+
"global_step": 45000,
|
| 540 |
+
"eval_loss": 9.493390083312988,
|
| 541 |
+
"eval_cosine_accuracy": 0.9352032542228699
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.595315732267452,
|
| 545 |
+
"global_step": 45500,
|
| 546 |
+
"eval_loss": 9.535852432250977,
|
| 547 |
+
"eval_cosine_accuracy": 0.9343876242637634
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 1.6128466743802812,
|
| 551 |
+
"global_step": 46000,
|
| 552 |
+
"eval_loss": 9.474373817443848,
|
| 553 |
+
"eval_cosine_accuracy": 0.9350672960281372
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 1.6303776164931103,
|
| 557 |
+
"global_step": 46500,
|
| 558 |
+
"eval_loss": 9.547622680664062,
|
| 559 |
+
"eval_cosine_accuracy": 0.9334360361099243
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 1.6479085586059394,
|
| 563 |
+
"global_step": 47000,
|
| 564 |
+
"eval_loss": 9.48145866394043,
|
| 565 |
+
"eval_cosine_accuracy": 0.9349313378334045
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 1.6654395007187688,
|
| 569 |
+
"global_step": 47500,
|
| 570 |
+
"eval_loss": 9.521085739135742,
|
| 571 |
+
"eval_cosine_accuracy": 0.9335266947746277
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 1.6829704428315977,
|
| 575 |
+
"global_step": 48000,
|
| 576 |
+
"eval_loss": 9.54494857788086,
|
| 577 |
+
"eval_cosine_accuracy": 0.934840738773346
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 1.700501384944427,
|
| 581 |
+
"global_step": 48500,
|
| 582 |
+
"eval_loss": 9.489590644836426,
|
| 583 |
+
"eval_cosine_accuracy": 0.935928225517273
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.718032327057256,
|
| 587 |
+
"global_step": 49000,
|
| 588 |
+
"eval_loss": 9.51984977722168,
|
| 589 |
+
"eval_cosine_accuracy": 0.933753252029419
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 1.7355632691700853,
|
| 593 |
+
"global_step": 49500,
|
| 594 |
+
"eval_loss": 9.579914093017578,
|
| 595 |
+
"eval_cosine_accuracy": 0.9317141771316528
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 1.7530942112829142,
|
| 599 |
+
"global_step": 50000,
|
| 600 |
+
"eval_loss": 9.662346839904785,
|
| 601 |
+
"eval_cosine_accuracy": 0.9311251044273376
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 1.7706251533957436,
|
| 605 |
+
"global_step": 50500,
|
| 606 |
+
"eval_loss": 9.592278480529785,
|
| 607 |
+
"eval_cosine_accuracy": 0.933390736579895
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 1.7881560955085727,
|
| 611 |
+
"global_step": 51000,
|
| 612 |
+
"eval_loss": 9.621561050415039,
|
| 613 |
+
"eval_cosine_accuracy": 0.932257890701294
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 1.8056870376214018,
|
| 617 |
+
"global_step": 51500,
|
| 618 |
+
"eval_loss": 9.6589937210083,
|
| 619 |
+
"eval_cosine_accuracy": 0.9350219964981079
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 1.823217979734231,
|
| 623 |
+
"global_step": 52000,
|
| 624 |
+
"eval_loss": 9.576985359191895,
|
| 625 |
+
"eval_cosine_accuracy": 0.9330735206604004
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.84074892184706,
|
| 629 |
+
"global_step": 52500,
|
| 630 |
+
"eval_loss": 9.682831764221191,
|
| 631 |
+
"eval_cosine_accuracy": 0.931034505367279
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 1.8582798639598892,
|
| 635 |
+
"global_step": 53000,
|
| 636 |
+
"eval_loss": 9.639060974121094,
|
| 637 |
+
"eval_cosine_accuracy": 0.9347047805786133
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 1.8758108060727183,
|
| 641 |
+
"global_step": 53500,
|
| 642 |
+
"eval_loss": 9.678288459777832,
|
| 643 |
+
"eval_cosine_accuracy": 0.9314423203468323
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 1.8933417481855475,
|
| 647 |
+
"global_step": 54000,
|
| 648 |
+
"eval_loss": 9.642071723937988,
|
| 649 |
+
"eval_cosine_accuracy": 0.9313969612121582
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 1.9108726902983766,
|
| 653 |
+
"global_step": 54500,
|
| 654 |
+
"eval_loss": 9.62327766418457,
|
| 655 |
+
"eval_cosine_accuracy": 0.9311704039573669
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 1.928403632411206,
|
| 659 |
+
"global_step": 55000,
|
| 660 |
+
"eval_loss": 9.659159660339355,
|
| 661 |
+
"eval_cosine_accuracy": 0.9341157078742981
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.9459345745240348,
|
| 665 |
+
"global_step": 55500,
|
| 666 |
+
"eval_loss": 9.677461624145508,
|
| 667 |
+
"eval_cosine_accuracy": 0.93248450756073
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.9634655166368642,
|
| 671 |
+
"global_step": 56000,
|
| 672 |
+
"eval_loss": 9.626531600952148,
|
| 673 |
+
"eval_cosine_accuracy": 0.9333454370498657
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 1.980996458749693,
|
| 677 |
+
"global_step": 56500,
|
| 678 |
+
"eval_loss": 9.674995422363281,
|
| 679 |
+
"eval_cosine_accuracy": 0.932121992111206
|
| 680 |
+
},
|
| 681 |
+
{
|
| 682 |
+
"epoch": 1.9985274008625225,
|
| 683 |
+
"global_step": 57000,
|
| 684 |
+
"eval_loss": 9.694042205810547,
|
| 685 |
+
"eval_cosine_accuracy": 0.9318501353263855
|
| 686 |
+
}
|
| 687 |
+
],
|
| 688 |
+
"current_epoch": 2,
|
| 689 |
+
"global_step": 57042
|
| 690 |
+
}
|
checkpoint-57042/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|