Training checkpoint - Epoch 4, Step 34368
Browse files- checkpoint-34368/1_Pooling/config.json +10 -0
- checkpoint-34368/config.json +25 -0
- checkpoint-34368/config_sentence_transformers.json +14 -0
- checkpoint-34368/model.safetensors +3 -0
- checkpoint-34368/modules.json +20 -0
- checkpoint-34368/optimizer.pt +3 -0
- checkpoint-34368/rng_state.pth +3 -0
- checkpoint-34368/scaler.pt +3 -0
- checkpoint-34368/scheduler.pt +3 -0
- checkpoint-34368/sentence_bert_config.json +4 -0
- checkpoint-34368/special_tokens_map.json +37 -0
- checkpoint-34368/tokenizer.json +0 -0
- checkpoint-34368/tokenizer_config.json +65 -0
- checkpoint-34368/trainer_state.json +1129 -0
- checkpoint-34368/training_args.bin +3 -0
- checkpoint-34368/training_metrics.json +414 -0
- checkpoint-34368/vocab.txt +0 -0
checkpoint-34368/1_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-34368/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-34368/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-34368/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:84a6902cff1b9ead24563f5cb1a8777a7a0a10fe67eb64bb087b15df250b0b81
|
| 3 |
+
size 90864192
|
checkpoint-34368/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-34368/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d1539daf68a06da744716a4d2016f3841b1641105cf11d8f9c77eadb2559f0b0
|
| 3 |
+
size 180607738
|
checkpoint-34368/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c0cae5f76925ad93d3a7ce5342bfff7c7f923659a33ac4262bde1a9c6bc52b6c
|
| 3 |
+
size 14244
|
checkpoint-34368/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a20782ce3892f7a1c7fe660a2c3154c28e2e368c586d4f2e0b09904cd1e3d407
|
| 3 |
+
size 988
|
checkpoint-34368/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1c4fd70635eb1c937cc668bfb6c74629b683c4c0b0391da40bd3af8128fdbaae
|
| 3 |
+
size 1064
|
checkpoint-34368/sentence_bert_config.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"max_seq_length": 256,
|
| 3 |
+
"do_lower_case": false
|
| 4 |
+
}
|
checkpoint-34368/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-34368/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-34368/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": 512,
|
| 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-34368/trainer_state.json
ADDED
|
@@ -0,0 +1,1129 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 4.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 34368,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.00011638733705772812,
|
| 14 |
+
"grad_norm": 3.3376758098602295,
|
| 15 |
+
"learning_rate": 0.0,
|
| 16 |
+
"loss": 0.8374,
|
| 17 |
+
"step": 1
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.05819366852886406,
|
| 21 |
+
"grad_norm": 1.1847290992736816,
|
| 22 |
+
"learning_rate": 1.7425212431614482e-06,
|
| 23 |
+
"loss": 0.5716,
|
| 24 |
+
"step": 500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.05819366852886406,
|
| 28 |
+
"eval_cosine_accuracy": 0.8337206840515137,
|
| 29 |
+
"eval_loss": 0.3265715539455414,
|
| 30 |
+
"eval_runtime": 22.7772,
|
| 31 |
+
"eval_samples_per_second": 415.591,
|
| 32 |
+
"eval_steps_per_second": 1.624,
|
| 33 |
+
"step": 500
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.11638733705772812,
|
| 37 |
+
"grad_norm": 0.9610480666160583,
|
| 38 |
+
"learning_rate": 3.4885345128622977e-06,
|
| 39 |
+
"loss": 0.2172,
|
| 40 |
+
"step": 1000
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.11638733705772812,
|
| 44 |
+
"eval_cosine_accuracy": 0.8280160427093506,
|
| 45 |
+
"eval_loss": 0.3617343008518219,
|
| 46 |
+
"eval_runtime": 24.0368,
|
| 47 |
+
"eval_samples_per_second": 393.812,
|
| 48 |
+
"eval_steps_per_second": 1.539,
|
| 49 |
+
"step": 1000
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.17458100558659218,
|
| 53 |
+
"grad_norm": 0.8891024589538574,
|
| 54 |
+
"learning_rate": 5.234547782563148e-06,
|
| 55 |
+
"loss": 0.1635,
|
| 56 |
+
"step": 1500
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.17458100558659218,
|
| 60 |
+
"eval_cosine_accuracy": 0.8382632732391357,
|
| 61 |
+
"eval_loss": 0.35951632261276245,
|
| 62 |
+
"eval_runtime": 23.0907,
|
| 63 |
+
"eval_samples_per_second": 409.949,
|
| 64 |
+
"eval_steps_per_second": 1.602,
|
| 65 |
+
"step": 1500
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.23277467411545624,
|
| 69 |
+
"grad_norm": 1.0580177307128906,
|
| 70 |
+
"learning_rate": 6.980561052263998e-06,
|
| 71 |
+
"loss": 0.1365,
|
| 72 |
+
"step": 2000
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.23277467411545624,
|
| 76 |
+
"eval_cosine_accuracy": 0.8358334898948669,
|
| 77 |
+
"eval_loss": 0.3711424171924591,
|
| 78 |
+
"eval_runtime": 21.9784,
|
| 79 |
+
"eval_samples_per_second": 430.695,
|
| 80 |
+
"eval_steps_per_second": 1.683,
|
| 81 |
+
"step": 2000
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.2909683426443203,
|
| 85 |
+
"grad_norm": 0.8555117249488831,
|
| 86 |
+
"learning_rate": 8.726574321964848e-06,
|
| 87 |
+
"loss": 0.1143,
|
| 88 |
+
"step": 2500
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.2909683426443203,
|
| 92 |
+
"eval_cosine_accuracy": 0.8282273411750793,
|
| 93 |
+
"eval_loss": 0.3904484808444977,
|
| 94 |
+
"eval_runtime": 21.699,
|
| 95 |
+
"eval_samples_per_second": 436.241,
|
| 96 |
+
"eval_steps_per_second": 1.705,
|
| 97 |
+
"step": 2500
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.34916201117318435,
|
| 101 |
+
"grad_norm": 0.7103874087333679,
|
| 102 |
+
"learning_rate": 1.0472587591665697e-05,
|
| 103 |
+
"loss": 0.1001,
|
| 104 |
+
"step": 3000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.34916201117318435,
|
| 108 |
+
"eval_cosine_accuracy": 0.822839617729187,
|
| 109 |
+
"eval_loss": 0.38800719380378723,
|
| 110 |
+
"eval_runtime": 22.7365,
|
| 111 |
+
"eval_samples_per_second": 416.336,
|
| 112 |
+
"eval_steps_per_second": 1.627,
|
| 113 |
+
"step": 3000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.4073556797020484,
|
| 117 |
+
"grad_norm": 0.8163154721260071,
|
| 118 |
+
"learning_rate": 1.2218600861366547e-05,
|
| 119 |
+
"loss": 0.0872,
|
| 120 |
+
"step": 3500
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.4073556797020484,
|
| 124 |
+
"eval_cosine_accuracy": 0.7903021574020386,
|
| 125 |
+
"eval_loss": 0.45464685559272766,
|
| 126 |
+
"eval_runtime": 21.8974,
|
| 127 |
+
"eval_samples_per_second": 432.289,
|
| 128 |
+
"eval_steps_per_second": 1.69,
|
| 129 |
+
"step": 3500
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.4655493482309125,
|
| 133 |
+
"grad_norm": 0.9644913077354431,
|
| 134 |
+
"learning_rate": 1.3964614131067397e-05,
|
| 135 |
+
"loss": 0.0788,
|
| 136 |
+
"step": 4000
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.4655493482309125,
|
| 140 |
+
"eval_cosine_accuracy": 0.7950559854507446,
|
| 141 |
+
"eval_loss": 0.4399067163467407,
|
| 142 |
+
"eval_runtime": 22.5922,
|
| 143 |
+
"eval_samples_per_second": 418.994,
|
| 144 |
+
"eval_steps_per_second": 1.638,
|
| 145 |
+
"step": 4000
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.5237430167597765,
|
| 149 |
+
"grad_norm": 0.7192422151565552,
|
| 150 |
+
"learning_rate": 1.5710627400768247e-05,
|
| 151 |
+
"loss": 0.0717,
|
| 152 |
+
"step": 4500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.5237430167597765,
|
| 156 |
+
"eval_cosine_accuracy": 0.7893513441085815,
|
| 157 |
+
"eval_loss": 0.44711923599243164,
|
| 158 |
+
"eval_runtime": 21.7249,
|
| 159 |
+
"eval_samples_per_second": 435.721,
|
| 160 |
+
"eval_steps_per_second": 1.703,
|
| 161 |
+
"step": 4500
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.5819366852886406,
|
| 165 |
+
"grad_norm": 0.7275687456130981,
|
| 166 |
+
"learning_rate": 1.7456640670469096e-05,
|
| 167 |
+
"loss": 0.0643,
|
| 168 |
+
"step": 5000
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 0.5819366852886406,
|
| 172 |
+
"eval_cosine_accuracy": 0.7972744703292847,
|
| 173 |
+
"eval_loss": 0.44583871960639954,
|
| 174 |
+
"eval_runtime": 22.8657,
|
| 175 |
+
"eval_samples_per_second": 413.983,
|
| 176 |
+
"eval_steps_per_second": 1.618,
|
| 177 |
+
"step": 5000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.6401303538175046,
|
| 181 |
+
"grad_norm": 0.6889352202415466,
|
| 182 |
+
"learning_rate": 1.9202653940169945e-05,
|
| 183 |
+
"loss": 0.06,
|
| 184 |
+
"step": 5500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.6401303538175046,
|
| 188 |
+
"eval_cosine_accuracy": 0.7760405540466309,
|
| 189 |
+
"eval_loss": 0.4766450822353363,
|
| 190 |
+
"eval_runtime": 22.1962,
|
| 191 |
+
"eval_samples_per_second": 426.47,
|
| 192 |
+
"eval_steps_per_second": 1.667,
|
| 193 |
+
"step": 5500
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.6983240223463687,
|
| 197 |
+
"grad_norm": 0.6319308280944824,
|
| 198 |
+
"learning_rate": 2.0948667209870795e-05,
|
| 199 |
+
"loss": 0.0571,
|
| 200 |
+
"step": 6000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.6983240223463687,
|
| 204 |
+
"eval_cosine_accuracy": 0.7615677118301392,
|
| 205 |
+
"eval_loss": 0.49710753560066223,
|
| 206 |
+
"eval_runtime": 22.6204,
|
| 207 |
+
"eval_samples_per_second": 418.472,
|
| 208 |
+
"eval_steps_per_second": 1.636,
|
| 209 |
+
"step": 6000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.7565176908752328,
|
| 213 |
+
"grad_norm": 0.4573739171028137,
|
| 214 |
+
"learning_rate": 2.2694680479571647e-05,
|
| 215 |
+
"loss": 0.0524,
|
| 216 |
+
"step": 6500
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.7565176908752328,
|
| 220 |
+
"eval_cosine_accuracy": 0.7761462330818176,
|
| 221 |
+
"eval_loss": 0.48617255687713623,
|
| 222 |
+
"eval_runtime": 21.8466,
|
| 223 |
+
"eval_samples_per_second": 433.294,
|
| 224 |
+
"eval_steps_per_second": 1.694,
|
| 225 |
+
"step": 6500
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.8147113594040968,
|
| 229 |
+
"grad_norm": 0.8050161600112915,
|
| 230 |
+
"learning_rate": 2.4440693749272496e-05,
|
| 231 |
+
"loss": 0.0488,
|
| 232 |
+
"step": 7000
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.8147113594040968,
|
| 236 |
+
"eval_cosine_accuracy": 0.7470948696136475,
|
| 237 |
+
"eval_loss": 0.514951765537262,
|
| 238 |
+
"eval_runtime": 22.7002,
|
| 239 |
+
"eval_samples_per_second": 417.001,
|
| 240 |
+
"eval_steps_per_second": 1.63,
|
| 241 |
+
"step": 7000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.8729050279329609,
|
| 245 |
+
"grad_norm": 0.5072457194328308,
|
| 246 |
+
"learning_rate": 2.6186707018973345e-05,
|
| 247 |
+
"loss": 0.0459,
|
| 248 |
+
"step": 7500
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.8729050279329609,
|
| 252 |
+
"eval_cosine_accuracy": 0.7597718238830566,
|
| 253 |
+
"eval_loss": 0.49496394395828247,
|
| 254 |
+
"eval_runtime": 21.7485,
|
| 255 |
+
"eval_samples_per_second": 435.247,
|
| 256 |
+
"eval_steps_per_second": 1.701,
|
| 257 |
+
"step": 7500
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.931098696461825,
|
| 261 |
+
"grad_norm": 0.5579398274421692,
|
| 262 |
+
"learning_rate": 2.7932720288674194e-05,
|
| 263 |
+
"loss": 0.0437,
|
| 264 |
+
"step": 8000
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.931098696461825,
|
| 268 |
+
"eval_cosine_accuracy": 0.7321994304656982,
|
| 269 |
+
"eval_loss": 0.5210782885551453,
|
| 270 |
+
"eval_runtime": 22.7473,
|
| 271 |
+
"eval_samples_per_second": 416.137,
|
| 272 |
+
"eval_steps_per_second": 1.627,
|
| 273 |
+
"step": 8000
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.9892923649906891,
|
| 277 |
+
"grad_norm": 0.5010477900505066,
|
| 278 |
+
"learning_rate": 2.9678733558375043e-05,
|
| 279 |
+
"loss": 0.0405,
|
| 280 |
+
"step": 8500
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.9892923649906891,
|
| 284 |
+
"eval_cosine_accuracy": 0.7376927733421326,
|
| 285 |
+
"eval_loss": 0.5197136402130127,
|
| 286 |
+
"eval_runtime": 22.3174,
|
| 287 |
+
"eval_samples_per_second": 424.154,
|
| 288 |
+
"eval_steps_per_second": 1.658,
|
| 289 |
+
"step": 8500
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 1.047486033519553,
|
| 293 |
+
"grad_norm": 0.6719731092453003,
|
| 294 |
+
"learning_rate": 2.998145925830976e-05,
|
| 295 |
+
"loss": 0.0374,
|
| 296 |
+
"step": 9000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 1.047486033519553,
|
| 300 |
+
"eval_cosine_accuracy": 0.7394887208938599,
|
| 301 |
+
"eval_loss": 0.5157160758972168,
|
| 302 |
+
"eval_runtime": 23.0569,
|
| 303 |
+
"eval_samples_per_second": 410.55,
|
| 304 |
+
"eval_steps_per_second": 1.605,
|
| 305 |
+
"step": 9000
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 1.105679702048417,
|
| 309 |
+
"grad_norm": 0.641225278377533,
|
| 310 |
+
"learning_rate": 2.9908447847348682e-05,
|
| 311 |
+
"loss": 0.035,
|
| 312 |
+
"step": 9500
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 1.105679702048417,
|
| 316 |
+
"eval_cosine_accuracy": 0.7379040718078613,
|
| 317 |
+
"eval_loss": 0.5262643694877625,
|
| 318 |
+
"eval_runtime": 21.8592,
|
| 319 |
+
"eval_samples_per_second": 433.045,
|
| 320 |
+
"eval_steps_per_second": 1.693,
|
| 321 |
+
"step": 9500
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 1.1638733705772812,
|
| 325 |
+
"grad_norm": 0.5788301825523376,
|
| 326 |
+
"learning_rate": 2.978000136191769e-05,
|
| 327 |
+
"loss": 0.0338,
|
| 328 |
+
"step": 10000
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 1.1638733705772812,
|
| 332 |
+
"eval_cosine_accuracy": 0.7258609533309937,
|
| 333 |
+
"eval_loss": 0.5438600778579712,
|
| 334 |
+
"eval_runtime": 23.5383,
|
| 335 |
+
"eval_samples_per_second": 402.154,
|
| 336 |
+
"eval_steps_per_second": 1.572,
|
| 337 |
+
"step": 10000
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 1.2220670391061452,
|
| 341 |
+
"grad_norm": 0.8140218257904053,
|
| 342 |
+
"learning_rate": 2.9596687387207934e-05,
|
| 343 |
+
"loss": 0.0334,
|
| 344 |
+
"step": 10500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 1.2220670391061452,
|
| 348 |
+
"eval_cosine_accuracy": 0.7293471097946167,
|
| 349 |
+
"eval_loss": 0.5367482900619507,
|
| 350 |
+
"eval_runtime": 22.2946,
|
| 351 |
+
"eval_samples_per_second": 424.587,
|
| 352 |
+
"eval_steps_per_second": 1.66,
|
| 353 |
+
"step": 10500
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.2802607076350094,
|
| 357 |
+
"grad_norm": 0.4544070363044739,
|
| 358 |
+
"learning_rate": 2.9359186435845607e-05,
|
| 359 |
+
"loss": 0.0298,
|
| 360 |
+
"step": 11000
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.2802607076350094,
|
| 364 |
+
"eval_cosine_accuracy": 0.7034650444984436,
|
| 365 |
+
"eval_loss": 0.5721879005432129,
|
| 366 |
+
"eval_runtime": 21.8905,
|
| 367 |
+
"eval_samples_per_second": 432.425,
|
| 368 |
+
"eval_steps_per_second": 1.69,
|
| 369 |
+
"step": 11000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 1.3384543761638734,
|
| 373 |
+
"grad_norm": 0.44614920020103455,
|
| 374 |
+
"learning_rate": 2.9068380177634623e-05,
|
| 375 |
+
"loss": 0.0298,
|
| 376 |
+
"step": 11500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 1.3384543761638734,
|
| 380 |
+
"eval_cosine_accuracy": 0.7343122959136963,
|
| 381 |
+
"eval_loss": 0.5377742648124695,
|
| 382 |
+
"eval_runtime": 21.8747,
|
| 383 |
+
"eval_samples_per_second": 432.737,
|
| 384 |
+
"eval_steps_per_second": 1.691,
|
| 385 |
+
"step": 11500
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 1.3966480446927374,
|
| 389 |
+
"grad_norm": 0.39302748441696167,
|
| 390 |
+
"learning_rate": 2.8726085537299306e-05,
|
| 391 |
+
"loss": 0.0293,
|
| 392 |
+
"step": 12000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 1.3966480446927374,
|
| 396 |
+
"eval_cosine_accuracy": 0.730192244052887,
|
| 397 |
+
"eval_loss": 0.5350662469863892,
|
| 398 |
+
"eval_runtime": 21.8411,
|
| 399 |
+
"eval_samples_per_second": 433.402,
|
| 400 |
+
"eval_steps_per_second": 1.694,
|
| 401 |
+
"step": 12000
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.4548417132216014,
|
| 405 |
+
"grad_norm": 0.7480568885803223,
|
| 406 |
+
"learning_rate": 2.8332201711072337e-05,
|
| 407 |
+
"loss": 0.0272,
|
| 408 |
+
"step": 12500
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.4548417132216014,
|
| 412 |
+
"eval_cosine_accuracy": 0.7192055583000183,
|
| 413 |
+
"eval_loss": 0.5529947876930237,
|
| 414 |
+
"eval_runtime": 21.7832,
|
| 415 |
+
"eval_samples_per_second": 434.554,
|
| 416 |
+
"eval_steps_per_second": 1.699,
|
| 417 |
+
"step": 12500
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 1.5130353817504656,
|
| 421 |
+
"grad_norm": 0.4840201139450073,
|
| 422 |
+
"learning_rate": 2.7888825031865706e-05,
|
| 423 |
+
"loss": 0.0269,
|
| 424 |
+
"step": 13000
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 1.5130353817504656,
|
| 428 |
+
"eval_cosine_accuracy": 0.7035706639289856,
|
| 429 |
+
"eval_loss": 0.5624330043792725,
|
| 430 |
+
"eval_runtime": 21.7886,
|
| 431 |
+
"eval_samples_per_second": 434.447,
|
| 432 |
+
"eval_steps_per_second": 1.698,
|
| 433 |
+
"step": 13000
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 1.5712290502793296,
|
| 437 |
+
"grad_norm": 0.5563199520111084,
|
| 438 |
+
"learning_rate": 2.7397601437632812e-05,
|
| 439 |
+
"loss": 0.0261,
|
| 440 |
+
"step": 13500
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 1.5712290502793296,
|
| 444 |
+
"eval_cosine_accuracy": 0.6944855451583862,
|
| 445 |
+
"eval_loss": 0.5798157453536987,
|
| 446 |
+
"eval_runtime": 21.8479,
|
| 447 |
+
"eval_samples_per_second": 433.268,
|
| 448 |
+
"eval_steps_per_second": 1.694,
|
| 449 |
+
"step": 13500
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 1.6294227188081938,
|
| 453 |
+
"grad_norm": 0.34846800565719604,
|
| 454 |
+
"learning_rate": 2.686035448742686e-05,
|
| 455 |
+
"loss": 0.0245,
|
| 456 |
+
"step": 14000
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 1.6294227188081938,
|
| 460 |
+
"eval_cosine_accuracy": 0.7114937901496887,
|
| 461 |
+
"eval_loss": 0.5619487166404724,
|
| 462 |
+
"eval_runtime": 21.951,
|
| 463 |
+
"eval_samples_per_second": 431.233,
|
| 464 |
+
"eval_steps_per_second": 1.686,
|
| 465 |
+
"step": 14000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 1.6876163873370578,
|
| 469 |
+
"grad_norm": 0.7336510419845581,
|
| 470 |
+
"learning_rate": 2.6279078591840833e-05,
|
| 471 |
+
"loss": 0.0249,
|
| 472 |
+
"step": 14500
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 1.6876163873370578,
|
| 476 |
+
"eval_cosine_accuracy": 0.6863511800765991,
|
| 477 |
+
"eval_loss": 0.5943849682807922,
|
| 478 |
+
"eval_runtime": 21.8986,
|
| 479 |
+
"eval_samples_per_second": 432.265,
|
| 480 |
+
"eval_steps_per_second": 1.69,
|
| 481 |
+
"step": 14500
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 1.7458100558659218,
|
| 485 |
+
"grad_norm": 0.3618803322315216,
|
| 486 |
+
"learning_rate": 2.565593160919884e-05,
|
| 487 |
+
"loss": 0.0239,
|
| 488 |
+
"step": 15000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 1.7458100558659218,
|
| 492 |
+
"eval_cosine_accuracy": 0.712444543838501,
|
| 493 |
+
"eval_loss": 0.5627889633178711,
|
| 494 |
+
"eval_runtime": 21.9628,
|
| 495 |
+
"eval_samples_per_second": 431.002,
|
| 496 |
+
"eval_steps_per_second": 1.685,
|
| 497 |
+
"step": 15000
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 1.8040037243947857,
|
| 501 |
+
"grad_norm": 0.5146603584289551,
|
| 502 |
+
"learning_rate": 2.4993226834983874e-05,
|
| 503 |
+
"loss": 0.0233,
|
| 504 |
+
"step": 15500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.8040037243947857,
|
| 508 |
+
"eval_cosine_accuracy": 0.698183000087738,
|
| 509 |
+
"eval_loss": 0.5584973692893982,
|
| 510 |
+
"eval_runtime": 21.7596,
|
| 511 |
+
"eval_samples_per_second": 435.026,
|
| 512 |
+
"eval_steps_per_second": 1.7,
|
| 513 |
+
"step": 15500
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.86219739292365,
|
| 517 |
+
"grad_norm": 0.2550230920314789,
|
| 518 |
+
"learning_rate": 2.4294859337799895e-05,
|
| 519 |
+
"loss": 0.0226,
|
| 520 |
+
"step": 16000
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.86219739292365,
|
| 524 |
+
"eval_cosine_accuracy": 0.7010352611541748,
|
| 525 |
+
"eval_loss": 0.5702577233314514,
|
| 526 |
+
"eval_runtime": 21.7371,
|
| 527 |
+
"eval_samples_per_second": 435.476,
|
| 528 |
+
"eval_steps_per_second": 1.702,
|
| 529 |
+
"step": 16000
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 1.920391061452514,
|
| 533 |
+
"grad_norm": 0.4447661340236664,
|
| 534 |
+
"learning_rate": 2.3560623452508205e-05,
|
| 535 |
+
"loss": 0.0219,
|
| 536 |
+
"step": 16500
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 1.920391061452514,
|
| 540 |
+
"eval_cosine_accuracy": 0.6801183223724365,
|
| 541 |
+
"eval_loss": 0.5965489745140076,
|
| 542 |
+
"eval_runtime": 21.8787,
|
| 543 |
+
"eval_samples_per_second": 432.658,
|
| 544 |
+
"eval_steps_per_second": 1.691,
|
| 545 |
+
"step": 16500
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 1.9785847299813781,
|
| 549 |
+
"grad_norm": 0.7168303728103638,
|
| 550 |
+
"learning_rate": 2.2794608144197627e-05,
|
| 551 |
+
"loss": 0.0214,
|
| 552 |
+
"step": 17000
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 1.9785847299813781,
|
| 556 |
+
"eval_cosine_accuracy": 0.7057891488075256,
|
| 557 |
+
"eval_loss": 0.5630908608436584,
|
| 558 |
+
"eval_runtime": 21.8174,
|
| 559 |
+
"eval_samples_per_second": 433.874,
|
| 560 |
+
"eval_steps_per_second": 1.696,
|
| 561 |
+
"step": 17000
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 2.036778398510242,
|
| 565 |
+
"grad_norm": 0.17964045703411102,
|
| 566 |
+
"learning_rate": 2.1999657075442132e-05,
|
| 567 |
+
"loss": 0.0187,
|
| 568 |
+
"step": 17500
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 2.036778398510242,
|
| 572 |
+
"eval_cosine_accuracy": 0.7000845074653625,
|
| 573 |
+
"eval_loss": 0.573241651058197,
|
| 574 |
+
"eval_runtime": 21.8664,
|
| 575 |
+
"eval_samples_per_second": 432.902,
|
| 576 |
+
"eval_steps_per_second": 1.692,
|
| 577 |
+
"step": 17500
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 2.094972067039106,
|
| 581 |
+
"grad_norm": 0.4394785463809967,
|
| 582 |
+
"learning_rate": 2.1180387119980224e-05,
|
| 583 |
+
"loss": 0.018,
|
| 584 |
+
"step": 18000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 2.094972067039106,
|
| 588 |
+
"eval_cosine_accuracy": 0.6810690760612488,
|
| 589 |
+
"eval_loss": 0.5863020420074463,
|
| 590 |
+
"eval_runtime": 21.9254,
|
| 591 |
+
"eval_samples_per_second": 431.737,
|
| 592 |
+
"eval_steps_per_second": 1.688,
|
| 593 |
+
"step": 18000
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 2.15316573556797,
|
| 597 |
+
"grad_norm": 0.47925424575805664,
|
| 598 |
+
"learning_rate": 2.0336557007607162e-05,
|
| 599 |
+
"loss": 0.018,
|
| 600 |
+
"step": 18500
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 2.15316573556797,
|
| 604 |
+
"eval_cosine_accuracy": 0.6927952766418457,
|
| 605 |
+
"eval_loss": 0.5871772766113281,
|
| 606 |
+
"eval_runtime": 21.775,
|
| 607 |
+
"eval_samples_per_second": 434.72,
|
| 608 |
+
"eval_steps_per_second": 1.699,
|
| 609 |
+
"step": 18500
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 2.211359404096834,
|
| 613 |
+
"grad_norm": 0.6263623833656311,
|
| 614 |
+
"learning_rate": 1.9472916106504752e-05,
|
| 615 |
+
"loss": 0.017,
|
| 616 |
+
"step": 19000
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 2.211359404096834,
|
| 620 |
+
"eval_cosine_accuracy": 0.6970208883285522,
|
| 621 |
+
"eval_loss": 0.5891428589820862,
|
| 622 |
+
"eval_runtime": 21.9388,
|
| 623 |
+
"eval_samples_per_second": 431.473,
|
| 624 |
+
"eval_steps_per_second": 1.687,
|
| 625 |
+
"step": 19000
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 2.2695530726256985,
|
| 629 |
+
"grad_norm": 0.2917712926864624,
|
| 630 |
+
"learning_rate": 1.8592670492688935e-05,
|
| 631 |
+
"loss": 0.0172,
|
| 632 |
+
"step": 19500
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 2.2695530726256985,
|
| 636 |
+
"eval_cosine_accuracy": 0.6833931803703308,
|
| 637 |
+
"eval_loss": 0.600773811340332,
|
| 638 |
+
"eval_runtime": 22.0018,
|
| 639 |
+
"eval_samples_per_second": 430.238,
|
| 640 |
+
"eval_steps_per_second": 1.682,
|
| 641 |
+
"step": 19500
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 2.3277467411545625,
|
| 645 |
+
"grad_norm": 0.4444028437137604,
|
| 646 |
+
"learning_rate": 1.7699087883502617e-05,
|
| 647 |
+
"loss": 0.0163,
|
| 648 |
+
"step": 20000
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 2.3277467411545625,
|
| 652 |
+
"eval_cosine_accuracy": 0.6958588361740112,
|
| 653 |
+
"eval_loss": 0.5844529271125793,
|
| 654 |
+
"eval_runtime": 21.9069,
|
| 655 |
+
"eval_samples_per_second": 432.101,
|
| 656 |
+
"eval_steps_per_second": 1.689,
|
| 657 |
+
"step": 20000
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 2.3859404096834265,
|
| 661 |
+
"grad_norm": 0.3189772963523865,
|
| 662 |
+
"learning_rate": 1.6795485506937118e-05,
|
| 663 |
+
"loss": 0.017,
|
| 664 |
+
"step": 20500
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 2.3859404096834265,
|
| 668 |
+
"eval_cosine_accuracy": 0.6862455010414124,
|
| 669 |
+
"eval_loss": 0.5948883295059204,
|
| 670 |
+
"eval_runtime": 21.6979,
|
| 671 |
+
"eval_samples_per_second": 436.263,
|
| 672 |
+
"eval_steps_per_second": 1.705,
|
| 673 |
+
"step": 20500
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 2.4441340782122905,
|
| 677 |
+
"grad_norm": 0.3407791256904602,
|
| 678 |
+
"learning_rate": 1.5885217787156163e-05,
|
| 679 |
+
"loss": 0.0162,
|
| 680 |
+
"step": 21000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 2.4441340782122905,
|
| 684 |
+
"eval_cosine_accuracy": 0.6829706430435181,
|
| 685 |
+
"eval_loss": 0.5971450805664062,
|
| 686 |
+
"eval_runtime": 21.9365,
|
| 687 |
+
"eval_samples_per_second": 431.519,
|
| 688 |
+
"eval_steps_per_second": 1.687,
|
| 689 |
+
"step": 21000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 2.5023277467411544,
|
| 693 |
+
"grad_norm": 0.6945669651031494,
|
| 694 |
+
"learning_rate": 1.4973492025972734e-05,
|
| 695 |
+
"loss": 0.0164,
|
| 696 |
+
"step": 21500
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 2.5023277467411544,
|
| 700 |
+
"eval_cosine_accuracy": 0.6735685467720032,
|
| 701 |
+
"eval_loss": 0.612638533115387,
|
| 702 |
+
"eval_runtime": 22.7793,
|
| 703 |
+
"eval_samples_per_second": 415.552,
|
| 704 |
+
"eval_steps_per_second": 1.624,
|
| 705 |
+
"step": 21500
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 2.560521415270019,
|
| 709 |
+
"grad_norm": 0.14824214577674866,
|
| 710 |
+
"learning_rate": 1.4060039725488674e-05,
|
| 711 |
+
"loss": 0.0157,
|
| 712 |
+
"step": 22000
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 2.560521415270019,
|
| 716 |
+
"eval_cosine_accuracy": 0.6825480461120605,
|
| 717 |
+
"eval_loss": 0.5996671319007874,
|
| 718 |
+
"eval_runtime": 21.8618,
|
| 719 |
+
"eval_samples_per_second": 432.993,
|
| 720 |
+
"eval_steps_per_second": 1.692,
|
| 721 |
+
"step": 22000
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 2.618715083798883,
|
| 725 |
+
"grad_norm": 0.4721280038356781,
|
| 726 |
+
"learning_rate": 1.3150076819830866e-05,
|
| 727 |
+
"loss": 0.0153,
|
| 728 |
+
"step": 22500
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 2.618715083798883,
|
| 732 |
+
"eval_cosine_accuracy": 0.6731460094451904,
|
| 733 |
+
"eval_loss": 0.6220535635948181,
|
| 734 |
+
"eval_runtime": 22.7089,
|
| 735 |
+
"eval_samples_per_second": 416.84,
|
| 736 |
+
"eval_steps_per_second": 1.629,
|
| 737 |
+
"step": 22500
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 2.676908752327747,
|
| 741 |
+
"grad_norm": 0.30308523774147034,
|
| 742 |
+
"learning_rate": 1.2246981345221882e-05,
|
| 743 |
+
"loss": 0.0152,
|
| 744 |
+
"step": 23000
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 2.676908752327747,
|
| 748 |
+
"eval_cosine_accuracy": 0.6774773001670837,
|
| 749 |
+
"eval_loss": 0.6126848459243774,
|
| 750 |
+
"eval_runtime": 22.0159,
|
| 751 |
+
"eval_samples_per_second": 429.961,
|
| 752 |
+
"eval_steps_per_second": 1.681,
|
| 753 |
+
"step": 23000
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 2.735102420856611,
|
| 757 |
+
"grad_norm": 0.3080209195613861,
|
| 758 |
+
"learning_rate": 1.1354105844063943e-05,
|
| 759 |
+
"loss": 0.0154,
|
| 760 |
+
"step": 23500
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 2.735102420856611,
|
| 764 |
+
"eval_cosine_accuracy": 0.6693429350852966,
|
| 765 |
+
"eval_loss": 0.6177366971969604,
|
| 766 |
+
"eval_runtime": 23.0053,
|
| 767 |
+
"eval_samples_per_second": 411.471,
|
| 768 |
+
"eval_steps_per_second": 1.608,
|
| 769 |
+
"step": 23500
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 2.793296089385475,
|
| 773 |
+
"grad_norm": 0.32632341980934143,
|
| 774 |
+
"learning_rate": 1.0474764919366047e-05,
|
| 775 |
+
"loss": 0.0151,
|
| 776 |
+
"step": 24000
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 2.793296089385475,
|
| 780 |
+
"eval_cosine_accuracy": 0.6803296208381653,
|
| 781 |
+
"eval_loss": 0.6116131544113159,
|
| 782 |
+
"eval_runtime": 21.8788,
|
| 783 |
+
"eval_samples_per_second": 432.656,
|
| 784 |
+
"eval_steps_per_second": 1.691,
|
| 785 |
+
"step": 24000
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 2.851489757914339,
|
| 789 |
+
"grad_norm": 0.4999234080314636,
|
| 790 |
+
"learning_rate": 9.612222930012653e-06,
|
| 791 |
+
"loss": 0.0152,
|
| 792 |
+
"step": 24500
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 2.851489757914339,
|
| 796 |
+
"eval_cosine_accuracy": 0.6775829195976257,
|
| 797 |
+
"eval_loss": 0.6035599112510681,
|
| 798 |
+
"eval_runtime": 22.7172,
|
| 799 |
+
"eval_samples_per_second": 416.688,
|
| 800 |
+
"eval_steps_per_second": 1.629,
|
| 801 |
+
"step": 24500
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 2.9096834264432028,
|
| 805 |
+
"grad_norm": 0.3499392569065094,
|
| 806 |
+
"learning_rate": 8.769681872552652e-06,
|
| 807 |
+
"loss": 0.0147,
|
| 808 |
+
"step": 25000
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 2.9096834264432028,
|
| 812 |
+
"eval_cosine_accuracy": 0.6763152480125427,
|
| 813 |
+
"eval_loss": 0.6071164011955261,
|
| 814 |
+
"eval_runtime": 22.2616,
|
| 815 |
+
"eval_samples_per_second": 425.216,
|
| 816 |
+
"eval_steps_per_second": 1.662,
|
| 817 |
+
"step": 25000
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 2.967877094972067,
|
| 821 |
+
"grad_norm": 0.3009015917778015,
|
| 822 |
+
"learning_rate": 7.951883199549091e-06,
|
| 823 |
+
"loss": 0.0147,
|
| 824 |
+
"step": 25500
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 2.967877094972067,
|
| 828 |
+
"eval_cosine_accuracy": 0.6777942180633545,
|
| 829 |
+
"eval_loss": 0.6033285856246948,
|
| 830 |
+
"eval_runtime": 22.9424,
|
| 831 |
+
"eval_samples_per_second": 412.599,
|
| 832 |
+
"eval_steps_per_second": 1.613,
|
| 833 |
+
"step": 25500
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 3.026070763500931,
|
| 837 |
+
"grad_norm": 0.3499734699726105,
|
| 838 |
+
"learning_rate": 7.160144581401225e-06,
|
| 839 |
+
"loss": 0.0139,
|
| 840 |
+
"step": 26000
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 3.026070763500931,
|
| 844 |
+
"eval_cosine_accuracy": 0.6765264868736267,
|
| 845 |
+
"eval_loss": 0.6075280904769897,
|
| 846 |
+
"eval_runtime": 22.0102,
|
| 847 |
+
"eval_samples_per_second": 430.074,
|
| 848 |
+
"eval_steps_per_second": 1.681,
|
| 849 |
+
"step": 26000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 3.084264432029795,
|
| 853 |
+
"grad_norm": 0.265242338180542,
|
| 854 |
+
"learning_rate": 6.395895884461541e-06,
|
| 855 |
+
"loss": 0.0124,
|
| 856 |
+
"step": 26500
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 3.084264432029795,
|
| 860 |
+
"eval_cosine_accuracy": 0.674836277961731,
|
| 861 |
+
"eval_loss": 0.6169347763061523,
|
| 862 |
+
"eval_runtime": 22.6711,
|
| 863 |
+
"eval_samples_per_second": 417.537,
|
| 864 |
+
"eval_steps_per_second": 1.632,
|
| 865 |
+
"step": 26500
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 3.142458100558659,
|
| 869 |
+
"grad_norm": 0.4871053993701935,
|
| 870 |
+
"learning_rate": 5.663588022894792e-06,
|
| 871 |
+
"loss": 0.0126,
|
| 872 |
+
"step": 27000
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 3.142458100558659,
|
| 876 |
+
"eval_cosine_accuracy": 0.674836277961731,
|
| 877 |
+
"eval_loss": 0.6152327656745911,
|
| 878 |
+
"eval_runtime": 22.233,
|
| 879 |
+
"eval_samples_per_second": 425.763,
|
| 880 |
+
"eval_steps_per_second": 1.664,
|
| 881 |
+
"step": 27000
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 3.200651769087523,
|
| 885 |
+
"grad_norm": 0.332655131816864,
|
| 886 |
+
"learning_rate": 4.965939527826877e-06,
|
| 887 |
+
"loss": 0.0127,
|
| 888 |
+
"step": 27500
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 3.200651769087523,
|
| 892 |
+
"eval_cosine_accuracy": 0.6781111359596252,
|
| 893 |
+
"eval_loss": 0.6097703576087952,
|
| 894 |
+
"eval_runtime": 23.0705,
|
| 895 |
+
"eval_samples_per_second": 410.307,
|
| 896 |
+
"eval_steps_per_second": 1.604,
|
| 897 |
+
"step": 27500
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 3.2588454376163876,
|
| 901 |
+
"grad_norm": 0.4277506172657013,
|
| 902 |
+
"learning_rate": 4.30554026514388e-06,
|
| 903 |
+
"loss": 0.0123,
|
| 904 |
+
"step": 28000
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 3.2588454376163876,
|
| 908 |
+
"eval_cosine_accuracy": 0.6747305989265442,
|
| 909 |
+
"eval_loss": 0.6135534048080444,
|
| 910 |
+
"eval_runtime": 21.8941,
|
| 911 |
+
"eval_samples_per_second": 432.354,
|
| 912 |
+
"eval_steps_per_second": 1.69,
|
| 913 |
+
"step": 28000
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 3.3170391061452515,
|
| 917 |
+
"grad_norm": 0.48300701379776,
|
| 918 |
+
"learning_rate": 3.6848418211872144e-06,
|
| 919 |
+
"loss": 0.0124,
|
| 920 |
+
"step": 28500
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 3.3170391061452515,
|
| 924 |
+
"eval_cosine_accuracy": 0.6752588152885437,
|
| 925 |
+
"eval_loss": 0.6127398610115051,
|
| 926 |
+
"eval_runtime": 21.9559,
|
| 927 |
+
"eval_samples_per_second": 431.137,
|
| 928 |
+
"eval_steps_per_second": 1.685,
|
| 929 |
+
"step": 28500
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 3.3752327746741155,
|
| 933 |
+
"grad_norm": 0.2515612244606018,
|
| 934 |
+
"learning_rate": 3.1061484017810333e-06,
|
| 935 |
+
"loss": 0.0125,
|
| 936 |
+
"step": 29000
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 3.3752327746741155,
|
| 940 |
+
"eval_cosine_accuracy": 0.6732516288757324,
|
| 941 |
+
"eval_loss": 0.6184883117675781,
|
| 942 |
+
"eval_runtime": 22.325,
|
| 943 |
+
"eval_samples_per_second": 424.008,
|
| 944 |
+
"eval_steps_per_second": 1.657,
|
| 945 |
+
"step": 29000
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 3.4334264432029795,
|
| 949 |
+
"grad_norm": 0.4237171411514282,
|
| 950 |
+
"learning_rate": 2.5716082783772475e-06,
|
| 951 |
+
"loss": 0.0128,
|
| 952 |
+
"step": 29500
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 3.4334264432029795,
|
| 956 |
+
"eval_cosine_accuracy": 0.6731460094451904,
|
| 957 |
+
"eval_loss": 0.6146544218063354,
|
| 958 |
+
"eval_runtime": 21.7418,
|
| 959 |
+
"eval_samples_per_second": 435.383,
|
| 960 |
+
"eval_steps_per_second": 1.702,
|
| 961 |
+
"step": 29500
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 3.4916201117318435,
|
| 965 |
+
"grad_norm": 0.3831160068511963,
|
| 966 |
+
"learning_rate": 2.083205813072427e-06,
|
| 967 |
+
"loss": 0.0117,
|
| 968 |
+
"step": 30000
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 3.4916201117318435,
|
| 972 |
+
"eval_cosine_accuracy": 0.6731460094451904,
|
| 973 |
+
"eval_loss": 0.6132456660270691,
|
| 974 |
+
"eval_runtime": 22.0358,
|
| 975 |
+
"eval_samples_per_second": 429.575,
|
| 976 |
+
"eval_steps_per_second": 1.679,
|
| 977 |
+
"step": 30000
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 3.5498137802607075,
|
| 981 |
+
"grad_norm": 0.2854015529155731,
|
| 982 |
+
"learning_rate": 1.643586023649546e-06,
|
| 983 |
+
"loss": 0.0121,
|
| 984 |
+
"step": 30500
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 3.5498137802607075,
|
| 988 |
+
"eval_cosine_accuracy": 0.6726177930831909,
|
| 989 |
+
"eval_loss": 0.6162088513374329,
|
| 990 |
+
"eval_runtime": 21.9437,
|
| 991 |
+
"eval_samples_per_second": 431.377,
|
| 992 |
+
"eval_steps_per_second": 1.686,
|
| 993 |
+
"step": 30500
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 3.6080074487895715,
|
| 997 |
+
"grad_norm": 0.26737916469573975,
|
| 998 |
+
"learning_rate": 1.252619444654759e-06,
|
| 999 |
+
"loss": 0.0121,
|
| 1000 |
+
"step": 31000
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 3.6080074487895715,
|
| 1004 |
+
"eval_cosine_accuracy": 0.6745193600654602,
|
| 1005 |
+
"eval_loss": 0.6143919229507446,
|
| 1006 |
+
"eval_runtime": 22.0114,
|
| 1007 |
+
"eval_samples_per_second": 430.049,
|
| 1008 |
+
"eval_steps_per_second": 1.681,
|
| 1009 |
+
"step": 31000
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 3.666201117318436,
|
| 1013 |
+
"grad_norm": 0.32189783453941345,
|
| 1014 |
+
"learning_rate": 9.126869783614116e-07,
|
| 1015 |
+
"loss": 0.0122,
|
| 1016 |
+
"step": 31500
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 3.666201117318436,
|
| 1020 |
+
"eval_cosine_accuracy": 0.6723008751869202,
|
| 1021 |
+
"eval_loss": 0.6197054982185364,
|
| 1022 |
+
"eval_runtime": 22.061,
|
| 1023 |
+
"eval_samples_per_second": 429.082,
|
| 1024 |
+
"eval_steps_per_second": 1.677,
|
| 1025 |
+
"step": 31500
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 3.7243947858473,
|
| 1029 |
+
"grad_norm": 0.4601638913154602,
|
| 1030 |
+
"learning_rate": 6.250505489394104e-07,
|
| 1031 |
+
"loss": 0.0124,
|
| 1032 |
+
"step": 32000
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 3.7243947858473,
|
| 1036 |
+
"eval_cosine_accuracy": 0.6715613603591919,
|
| 1037 |
+
"eval_loss": 0.619126558303833,
|
| 1038 |
+
"eval_runtime": 22.0063,
|
| 1039 |
+
"eval_samples_per_second": 430.15,
|
| 1040 |
+
"eval_steps_per_second": 1.681,
|
| 1041 |
+
"step": 32000
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 3.782588454376164,
|
| 1045 |
+
"grad_norm": 0.4083310663700104,
|
| 1046 |
+
"learning_rate": 3.911926198100924e-07,
|
| 1047 |
+
"loss": 0.0124,
|
| 1048 |
+
"step": 32500
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 3.782588454376164,
|
| 1052 |
+
"eval_cosine_accuracy": 0.6720895767211914,
|
| 1053 |
+
"eval_loss": 0.6180660128593445,
|
| 1054 |
+
"eval_runtime": 22.0652,
|
| 1055 |
+
"eval_samples_per_second": 429.002,
|
| 1056 |
+
"eval_steps_per_second": 1.677,
|
| 1057 |
+
"step": 32500
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 3.840782122905028,
|
| 1061 |
+
"grad_norm": 0.5940556526184082,
|
| 1062 |
+
"learning_rate": 2.1104432118265905e-07,
|
| 1063 |
+
"loss": 0.012,
|
| 1064 |
+
"step": 33000
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 3.840782122905028,
|
| 1068 |
+
"eval_cosine_accuracy": 0.6726177930831909,
|
| 1069 |
+
"eval_loss": 0.6176912188529968,
|
| 1070 |
+
"eval_runtime": 22.1115,
|
| 1071 |
+
"eval_samples_per_second": 428.102,
|
| 1072 |
+
"eval_steps_per_second": 1.673,
|
| 1073 |
+
"step": 33000
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 3.898975791433892,
|
| 1077 |
+
"grad_norm": 0.3067028820514679,
|
| 1078 |
+
"learning_rate": 8.57967527569431e-08,
|
| 1079 |
+
"loss": 0.0119,
|
| 1080 |
+
"step": 33500
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 3.898975791433892,
|
| 1084 |
+
"eval_cosine_accuracy": 0.6723008751869202,
|
| 1085 |
+
"eval_loss": 0.6178871989250183,
|
| 1086 |
+
"eval_runtime": 22.1297,
|
| 1087 |
+
"eval_samples_per_second": 427.751,
|
| 1088 |
+
"eval_steps_per_second": 1.672,
|
| 1089 |
+
"step": 33500
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 3.9571694599627563,
|
| 1093 |
+
"grad_norm": 0.43891850113868713,
|
| 1094 |
+
"learning_rate": 1.591486845612389e-08,
|
| 1095 |
+
"loss": 0.0126,
|
| 1096 |
+
"step": 34000
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 3.9571694599627563,
|
| 1100 |
+
"eval_cosine_accuracy": 0.6720895767211914,
|
| 1101 |
+
"eval_loss": 0.6180205345153809,
|
| 1102 |
+
"eval_runtime": 21.7921,
|
| 1103 |
+
"eval_samples_per_second": 434.378,
|
| 1104 |
+
"eval_steps_per_second": 1.698,
|
| 1105 |
+
"step": 34000
|
| 1106 |
+
}
|
| 1107 |
+
],
|
| 1108 |
+
"logging_steps": 500,
|
| 1109 |
+
"max_steps": 34368,
|
| 1110 |
+
"num_input_tokens_seen": 0,
|
| 1111 |
+
"num_train_epochs": 4,
|
| 1112 |
+
"save_steps": 500,
|
| 1113 |
+
"stateful_callbacks": {
|
| 1114 |
+
"TrainerControl": {
|
| 1115 |
+
"args": {
|
| 1116 |
+
"should_epoch_stop": false,
|
| 1117 |
+
"should_evaluate": false,
|
| 1118 |
+
"should_log": false,
|
| 1119 |
+
"should_save": true,
|
| 1120 |
+
"should_training_stop": true
|
| 1121 |
+
},
|
| 1122 |
+
"attributes": {}
|
| 1123 |
+
}
|
| 1124 |
+
},
|
| 1125 |
+
"total_flos": 0.0,
|
| 1126 |
+
"train_batch_size": 256,
|
| 1127 |
+
"trial_name": null,
|
| 1128 |
+
"trial_params": null
|
| 1129 |
+
}
|
checkpoint-34368/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:948831375bed1d29c3794b486159432f0dfcf3c8ec9b8b77710140bfb9013d6f
|
| 3 |
+
size 5624
|
checkpoint-34368/training_metrics.json
ADDED
|
@@ -0,0 +1,414 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metrics_history": [
|
| 3 |
+
{
|
| 4 |
+
"epoch": 0.05819366852886406,
|
| 5 |
+
"global_step": 500,
|
| 6 |
+
"eval_loss": 0.3265715539455414,
|
| 7 |
+
"eval_cosine_accuracy": 0.8337206840515137
|
| 8 |
+
},
|
| 9 |
+
{
|
| 10 |
+
"epoch": 0.11638733705772812,
|
| 11 |
+
"global_step": 1000,
|
| 12 |
+
"eval_loss": 0.3617343008518219,
|
| 13 |
+
"eval_cosine_accuracy": 0.8280160427093506
|
| 14 |
+
},
|
| 15 |
+
{
|
| 16 |
+
"epoch": 0.17458100558659218,
|
| 17 |
+
"global_step": 1500,
|
| 18 |
+
"eval_loss": 0.35951632261276245,
|
| 19 |
+
"eval_cosine_accuracy": 0.8382632732391357
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"epoch": 0.23277467411545624,
|
| 23 |
+
"global_step": 2000,
|
| 24 |
+
"eval_loss": 0.3711424171924591,
|
| 25 |
+
"eval_cosine_accuracy": 0.8358334898948669
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.2909683426443203,
|
| 29 |
+
"global_step": 2500,
|
| 30 |
+
"eval_loss": 0.3904484808444977,
|
| 31 |
+
"eval_cosine_accuracy": 0.8282273411750793
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.34916201117318435,
|
| 35 |
+
"global_step": 3000,
|
| 36 |
+
"eval_loss": 0.38800719380378723,
|
| 37 |
+
"eval_cosine_accuracy": 0.822839617729187
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.4073556797020484,
|
| 41 |
+
"global_step": 3500,
|
| 42 |
+
"eval_loss": 0.45464685559272766,
|
| 43 |
+
"eval_cosine_accuracy": 0.7903021574020386
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 0.4655493482309125,
|
| 47 |
+
"global_step": 4000,
|
| 48 |
+
"eval_loss": 0.4399067163467407,
|
| 49 |
+
"eval_cosine_accuracy": 0.7950559854507446
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.5237430167597765,
|
| 53 |
+
"global_step": 4500,
|
| 54 |
+
"eval_loss": 0.44711923599243164,
|
| 55 |
+
"eval_cosine_accuracy": 0.7893513441085815
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.5819366852886406,
|
| 59 |
+
"global_step": 5000,
|
| 60 |
+
"eval_loss": 0.44583871960639954,
|
| 61 |
+
"eval_cosine_accuracy": 0.7972744703292847
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.6401303538175046,
|
| 65 |
+
"global_step": 5500,
|
| 66 |
+
"eval_loss": 0.4766450822353363,
|
| 67 |
+
"eval_cosine_accuracy": 0.7760405540466309
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.6983240223463687,
|
| 71 |
+
"global_step": 6000,
|
| 72 |
+
"eval_loss": 0.49710753560066223,
|
| 73 |
+
"eval_cosine_accuracy": 0.7615677118301392
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.7565176908752328,
|
| 77 |
+
"global_step": 6500,
|
| 78 |
+
"eval_loss": 0.48617255687713623,
|
| 79 |
+
"eval_cosine_accuracy": 0.7761462330818176
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.8147113594040968,
|
| 83 |
+
"global_step": 7000,
|
| 84 |
+
"eval_loss": 0.514951765537262,
|
| 85 |
+
"eval_cosine_accuracy": 0.7470948696136475
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.8729050279329609,
|
| 89 |
+
"global_step": 7500,
|
| 90 |
+
"eval_loss": 0.49496394395828247,
|
| 91 |
+
"eval_cosine_accuracy": 0.7597718238830566
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.931098696461825,
|
| 95 |
+
"global_step": 8000,
|
| 96 |
+
"eval_loss": 0.5210782885551453,
|
| 97 |
+
"eval_cosine_accuracy": 0.7321994304656982
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.9892923649906891,
|
| 101 |
+
"global_step": 8500,
|
| 102 |
+
"eval_loss": 0.5197136402130127,
|
| 103 |
+
"eval_cosine_accuracy": 0.7376927733421326
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 1.047486033519553,
|
| 107 |
+
"global_step": 9000,
|
| 108 |
+
"eval_loss": 0.5157160758972168,
|
| 109 |
+
"eval_cosine_accuracy": 0.7394887208938599
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.105679702048417,
|
| 113 |
+
"global_step": 9500,
|
| 114 |
+
"eval_loss": 0.5262643694877625,
|
| 115 |
+
"eval_cosine_accuracy": 0.7379040718078613
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 1.1638733705772812,
|
| 119 |
+
"global_step": 10000,
|
| 120 |
+
"eval_loss": 0.5438600778579712,
|
| 121 |
+
"eval_cosine_accuracy": 0.7258609533309937
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 1.2220670391061452,
|
| 125 |
+
"global_step": 10500,
|
| 126 |
+
"eval_loss": 0.5367482900619507,
|
| 127 |
+
"eval_cosine_accuracy": 0.7293471097946167
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 1.2802607076350094,
|
| 131 |
+
"global_step": 11000,
|
| 132 |
+
"eval_loss": 0.5721879005432129,
|
| 133 |
+
"eval_cosine_accuracy": 0.7034650444984436
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 1.3384543761638734,
|
| 137 |
+
"global_step": 11500,
|
| 138 |
+
"eval_loss": 0.5377742648124695,
|
| 139 |
+
"eval_cosine_accuracy": 0.7343122959136963
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 1.3966480446927374,
|
| 143 |
+
"global_step": 12000,
|
| 144 |
+
"eval_loss": 0.5350662469863892,
|
| 145 |
+
"eval_cosine_accuracy": 0.730192244052887
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 1.4548417132216014,
|
| 149 |
+
"global_step": 12500,
|
| 150 |
+
"eval_loss": 0.5529947876930237,
|
| 151 |
+
"eval_cosine_accuracy": 0.7192055583000183
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.5130353817504656,
|
| 155 |
+
"global_step": 13000,
|
| 156 |
+
"eval_loss": 0.5624330043792725,
|
| 157 |
+
"eval_cosine_accuracy": 0.7035706639289856
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 1.5712290502793296,
|
| 161 |
+
"global_step": 13500,
|
| 162 |
+
"eval_loss": 0.5798157453536987,
|
| 163 |
+
"eval_cosine_accuracy": 0.6944855451583862
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 1.6294227188081938,
|
| 167 |
+
"global_step": 14000,
|
| 168 |
+
"eval_loss": 0.5619487166404724,
|
| 169 |
+
"eval_cosine_accuracy": 0.7114937901496887
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 1.6876163873370578,
|
| 173 |
+
"global_step": 14500,
|
| 174 |
+
"eval_loss": 0.5943849682807922,
|
| 175 |
+
"eval_cosine_accuracy": 0.6863511800765991
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 1.7458100558659218,
|
| 179 |
+
"global_step": 15000,
|
| 180 |
+
"eval_loss": 0.5627889633178711,
|
| 181 |
+
"eval_cosine_accuracy": 0.712444543838501
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 1.8040037243947857,
|
| 185 |
+
"global_step": 15500,
|
| 186 |
+
"eval_loss": 0.5584973692893982,
|
| 187 |
+
"eval_cosine_accuracy": 0.698183000087738
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 1.86219739292365,
|
| 191 |
+
"global_step": 16000,
|
| 192 |
+
"eval_loss": 0.5702577233314514,
|
| 193 |
+
"eval_cosine_accuracy": 0.7010352611541748
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 1.920391061452514,
|
| 197 |
+
"global_step": 16500,
|
| 198 |
+
"eval_loss": 0.5965489745140076,
|
| 199 |
+
"eval_cosine_accuracy": 0.6801183223724365
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.9785847299813781,
|
| 203 |
+
"global_step": 17000,
|
| 204 |
+
"eval_loss": 0.5630908608436584,
|
| 205 |
+
"eval_cosine_accuracy": 0.7057891488075256
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 2.036778398510242,
|
| 209 |
+
"global_step": 17500,
|
| 210 |
+
"eval_loss": 0.573241651058197,
|
| 211 |
+
"eval_cosine_accuracy": 0.7000845074653625
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 2.094972067039106,
|
| 215 |
+
"global_step": 18000,
|
| 216 |
+
"eval_loss": 0.5863020420074463,
|
| 217 |
+
"eval_cosine_accuracy": 0.6810690760612488
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 2.15316573556797,
|
| 221 |
+
"global_step": 18500,
|
| 222 |
+
"eval_loss": 0.5871772766113281,
|
| 223 |
+
"eval_cosine_accuracy": 0.6927952766418457
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 2.211359404096834,
|
| 227 |
+
"global_step": 19000,
|
| 228 |
+
"eval_loss": 0.5891428589820862,
|
| 229 |
+
"eval_cosine_accuracy": 0.6970208883285522
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 2.2695530726256985,
|
| 233 |
+
"global_step": 19500,
|
| 234 |
+
"eval_loss": 0.600773811340332,
|
| 235 |
+
"eval_cosine_accuracy": 0.6833931803703308
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 2.3277467411545625,
|
| 239 |
+
"global_step": 20000,
|
| 240 |
+
"eval_loss": 0.5844529271125793,
|
| 241 |
+
"eval_cosine_accuracy": 0.6958588361740112
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 2.3859404096834265,
|
| 245 |
+
"global_step": 20500,
|
| 246 |
+
"eval_loss": 0.5948883295059204,
|
| 247 |
+
"eval_cosine_accuracy": 0.6862455010414124
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 2.4441340782122905,
|
| 251 |
+
"global_step": 21000,
|
| 252 |
+
"eval_loss": 0.5971450805664062,
|
| 253 |
+
"eval_cosine_accuracy": 0.6829706430435181
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 2.5023277467411544,
|
| 257 |
+
"global_step": 21500,
|
| 258 |
+
"eval_loss": 0.612638533115387,
|
| 259 |
+
"eval_cosine_accuracy": 0.6735685467720032
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 2.560521415270019,
|
| 263 |
+
"global_step": 22000,
|
| 264 |
+
"eval_loss": 0.5996671319007874,
|
| 265 |
+
"eval_cosine_accuracy": 0.6825480461120605
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 2.618715083798883,
|
| 269 |
+
"global_step": 22500,
|
| 270 |
+
"eval_loss": 0.6220535635948181,
|
| 271 |
+
"eval_cosine_accuracy": 0.6731460094451904
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 2.676908752327747,
|
| 275 |
+
"global_step": 23000,
|
| 276 |
+
"eval_loss": 0.6126848459243774,
|
| 277 |
+
"eval_cosine_accuracy": 0.6774773001670837
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 2.735102420856611,
|
| 281 |
+
"global_step": 23500,
|
| 282 |
+
"eval_loss": 0.6177366971969604,
|
| 283 |
+
"eval_cosine_accuracy": 0.6693429350852966
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 2.793296089385475,
|
| 287 |
+
"global_step": 24000,
|
| 288 |
+
"eval_loss": 0.6116131544113159,
|
| 289 |
+
"eval_cosine_accuracy": 0.6803296208381653
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 2.851489757914339,
|
| 293 |
+
"global_step": 24500,
|
| 294 |
+
"eval_loss": 0.6035599112510681,
|
| 295 |
+
"eval_cosine_accuracy": 0.6775829195976257
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 2.9096834264432028,
|
| 299 |
+
"global_step": 25000,
|
| 300 |
+
"eval_loss": 0.6071164011955261,
|
| 301 |
+
"eval_cosine_accuracy": 0.6763152480125427
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 2.967877094972067,
|
| 305 |
+
"global_step": 25500,
|
| 306 |
+
"eval_loss": 0.6033285856246948,
|
| 307 |
+
"eval_cosine_accuracy": 0.6777942180633545
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 3.026070763500931,
|
| 311 |
+
"global_step": 26000,
|
| 312 |
+
"eval_loss": 0.6075280904769897,
|
| 313 |
+
"eval_cosine_accuracy": 0.6765264868736267
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 3.084264432029795,
|
| 317 |
+
"global_step": 26500,
|
| 318 |
+
"eval_loss": 0.6169347763061523,
|
| 319 |
+
"eval_cosine_accuracy": 0.674836277961731
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 3.142458100558659,
|
| 323 |
+
"global_step": 27000,
|
| 324 |
+
"eval_loss": 0.6152327656745911,
|
| 325 |
+
"eval_cosine_accuracy": 0.674836277961731
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 3.200651769087523,
|
| 329 |
+
"global_step": 27500,
|
| 330 |
+
"eval_loss": 0.6097703576087952,
|
| 331 |
+
"eval_cosine_accuracy": 0.6781111359596252
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 3.2588454376163876,
|
| 335 |
+
"global_step": 28000,
|
| 336 |
+
"eval_loss": 0.6135534048080444,
|
| 337 |
+
"eval_cosine_accuracy": 0.6747305989265442
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 3.3170391061452515,
|
| 341 |
+
"global_step": 28500,
|
| 342 |
+
"eval_loss": 0.6127398610115051,
|
| 343 |
+
"eval_cosine_accuracy": 0.6752588152885437
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 3.3752327746741155,
|
| 347 |
+
"global_step": 29000,
|
| 348 |
+
"eval_loss": 0.6184883117675781,
|
| 349 |
+
"eval_cosine_accuracy": 0.6732516288757324
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 3.4334264432029795,
|
| 353 |
+
"global_step": 29500,
|
| 354 |
+
"eval_loss": 0.6146544218063354,
|
| 355 |
+
"eval_cosine_accuracy": 0.6731460094451904
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 3.4916201117318435,
|
| 359 |
+
"global_step": 30000,
|
| 360 |
+
"eval_loss": 0.6132456660270691,
|
| 361 |
+
"eval_cosine_accuracy": 0.6731460094451904
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 3.5498137802607075,
|
| 365 |
+
"global_step": 30500,
|
| 366 |
+
"eval_loss": 0.6162088513374329,
|
| 367 |
+
"eval_cosine_accuracy": 0.6726177930831909
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 3.6080074487895715,
|
| 371 |
+
"global_step": 31000,
|
| 372 |
+
"eval_loss": 0.6143919229507446,
|
| 373 |
+
"eval_cosine_accuracy": 0.6745193600654602
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 3.666201117318436,
|
| 377 |
+
"global_step": 31500,
|
| 378 |
+
"eval_loss": 0.6197054982185364,
|
| 379 |
+
"eval_cosine_accuracy": 0.6723008751869202
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 3.7243947858473,
|
| 383 |
+
"global_step": 32000,
|
| 384 |
+
"eval_loss": 0.619126558303833,
|
| 385 |
+
"eval_cosine_accuracy": 0.6715613603591919
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 3.782588454376164,
|
| 389 |
+
"global_step": 32500,
|
| 390 |
+
"eval_loss": 0.6180660128593445,
|
| 391 |
+
"eval_cosine_accuracy": 0.6720895767211914
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 3.840782122905028,
|
| 395 |
+
"global_step": 33000,
|
| 396 |
+
"eval_loss": 0.6176912188529968,
|
| 397 |
+
"eval_cosine_accuracy": 0.6726177930831909
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 3.898975791433892,
|
| 401 |
+
"global_step": 33500,
|
| 402 |
+
"eval_loss": 0.6178871989250183,
|
| 403 |
+
"eval_cosine_accuracy": 0.6723008751869202
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 3.9571694599627563,
|
| 407 |
+
"global_step": 34000,
|
| 408 |
+
"eval_loss": 0.6180205345153809,
|
| 409 |
+
"eval_cosine_accuracy": 0.6720895767211914
|
| 410 |
+
}
|
| 411 |
+
],
|
| 412 |
+
"current_epoch": 4,
|
| 413 |
+
"global_step": 34368
|
| 414 |
+
}
|
checkpoint-34368/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|