Upload 6 files
Browse files- config.json +25 -0
- model.safetensors +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +16 -0
- trainer_state.json +1432 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"MPNetForMaskedLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"dtype": "float16",
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"layer_norm_eps": 1e-05,
|
| 15 |
+
"max_position_embeddings": 514,
|
| 16 |
+
"model_type": "mpnet",
|
| 17 |
+
"num_attention_heads": 12,
|
| 18 |
+
"num_hidden_layers": 12,
|
| 19 |
+
"pad_token_id": 1,
|
| 20 |
+
"relative_attention_num_buckets": 32,
|
| 21 |
+
"tie_word_embeddings": true,
|
| 22 |
+
"transformers_version": "5.0.0",
|
| 23 |
+
"use_cache": false,
|
| 24 |
+
"vocab_size": 30527
|
| 25 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:345e644416dff53ce516d961eeb96dea467ada38a7172d8dbefd5f62144db8bb
|
| 3 |
+
size 266011004
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "<s>",
|
| 4 |
+
"cls_token": "<s>",
|
| 5 |
+
"do_lower_case": true,
|
| 6 |
+
"eos_token": "</s>",
|
| 7 |
+
"is_local": true,
|
| 8 |
+
"mask_token": "<mask>",
|
| 9 |
+
"model_max_length": 512,
|
| 10 |
+
"pad_token": "<pad>",
|
| 11 |
+
"sep_token": "</s>",
|
| 12 |
+
"strip_accents": null,
|
| 13 |
+
"tokenize_chinese_chars": true,
|
| 14 |
+
"tokenizer_class": "MPNetTokenizer",
|
| 15 |
+
"unk_token": "[UNK]"
|
| 16 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1432 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 16.515276630883566,
|
| 6 |
+
"eval_steps": 2000,
|
| 7 |
+
"global_step": 120000,
|
| 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,
|
| 14 |
+
"eval_accuracy": 1.2790688050781904e-05,
|
| 15 |
+
"eval_loss": 115.5,
|
| 16 |
+
"eval_runtime": 200.9168,
|
| 17 |
+
"eval_samples_per_second": 8229.121,
|
| 18 |
+
"eval_steps_per_second": 16.076,
|
| 19 |
+
"step": 0
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"epoch": 0.13762730525736305,
|
| 23 |
+
"grad_norm": 26.74563980102539,
|
| 24 |
+
"learning_rate": 2.9639999999999997e-05,
|
| 25 |
+
"loss": 141.26440625,
|
| 26 |
+
"step": 1000
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.2752546105147261,
|
| 30 |
+
"grad_norm": 22.885421752929688,
|
| 31 |
+
"learning_rate": 5.964e-05,
|
| 32 |
+
"loss": 59.12350390625,
|
| 33 |
+
"step": 2000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.2752546105147261,
|
| 37 |
+
"eval_accuracy": 0.676248915563685,
|
| 38 |
+
"eval_loss": 12.453125,
|
| 39 |
+
"eval_runtime": 168.1994,
|
| 40 |
+
"eval_samples_per_second": 9829.817,
|
| 41 |
+
"eval_steps_per_second": 19.203,
|
| 42 |
+
"step": 2000
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.41288191577208916,
|
| 46 |
+
"grad_norm": 21.04855728149414,
|
| 47 |
+
"learning_rate": 8.957999999999998e-05,
|
| 48 |
+
"loss": 50.17753125,
|
| 49 |
+
"step": 3000
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.5505092210294522,
|
| 53 |
+
"grad_norm": 20.75350570678711,
|
| 54 |
+
"learning_rate": 0.00011957999999999999,
|
| 55 |
+
"loss": 46.33467578125,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.5505092210294522,
|
| 60 |
+
"eval_accuracy": 0.7075093226563655,
|
| 61 |
+
"eval_loss": 10.6796875,
|
| 62 |
+
"eval_runtime": 167.9148,
|
| 63 |
+
"eval_samples_per_second": 9846.476,
|
| 64 |
+
"eval_steps_per_second": 19.236,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.6881365262868153,
|
| 69 |
+
"grad_norm": 19.514423370361328,
|
| 70 |
+
"learning_rate": 0.00014948999999999998,
|
| 71 |
+
"loss": 44.1445078125,
|
| 72 |
+
"step": 5000
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.8257638315441783,
|
| 76 |
+
"grad_norm": 18.586366653442383,
|
| 77 |
+
"learning_rate": 0.00017949,
|
| 78 |
+
"loss": 42.90046875,
|
| 79 |
+
"step": 6000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.8257638315441783,
|
| 83 |
+
"eval_accuracy": 0.7171687154928088,
|
| 84 |
+
"eval_loss": 10.140625,
|
| 85 |
+
"eval_runtime": 169.3136,
|
| 86 |
+
"eval_samples_per_second": 9765.126,
|
| 87 |
+
"eval_steps_per_second": 19.077,
|
| 88 |
+
"step": 6000
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.9633911368015414,
|
| 92 |
+
"grad_norm": 17.66230010986328,
|
| 93 |
+
"learning_rate": 0.00020946,
|
| 94 |
+
"loss": 41.96175390625,
|
| 95 |
+
"step": 7000
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.1010184420589044,
|
| 99 |
+
"grad_norm": 19.431285858154297,
|
| 100 |
+
"learning_rate": 0.0002394,
|
| 101 |
+
"loss": 41.42766015625,
|
| 102 |
+
"step": 8000
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.1010184420589044,
|
| 106 |
+
"eval_accuracy": 0.7211219507641113,
|
| 107 |
+
"eval_loss": 9.921875,
|
| 108 |
+
"eval_runtime": 170.601,
|
| 109 |
+
"eval_samples_per_second": 9691.437,
|
| 110 |
+
"eval_steps_per_second": 18.933,
|
| 111 |
+
"step": 8000
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 1.2386457473162675,
|
| 115 |
+
"grad_norm": 19.449247360229492,
|
| 116 |
+
"learning_rate": 0.00026933999999999997,
|
| 117 |
+
"loss": 41.1454921875,
|
| 118 |
+
"step": 9000
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 1.3762730525736306,
|
| 122 |
+
"grad_norm": 21.419654846191406,
|
| 123 |
+
"learning_rate": 0.00029934,
|
| 124 |
+
"loss": 40.98086328125,
|
| 125 |
+
"step": 10000
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 1.3762730525736306,
|
| 129 |
+
"eval_accuracy": 0.7225334877311377,
|
| 130 |
+
"eval_loss": 9.8359375,
|
| 131 |
+
"eval_runtime": 167.923,
|
| 132 |
+
"eval_samples_per_second": 9845.998,
|
| 133 |
+
"eval_steps_per_second": 19.235,
|
| 134 |
+
"step": 10000
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.5139003578309937,
|
| 138 |
+
"grad_norm": 17.391115188598633,
|
| 139 |
+
"learning_rate": 0.0003,
|
| 140 |
+
"loss": 40.79280078125,
|
| 141 |
+
"step": 11000
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 1.6515276630883569,
|
| 145 |
+
"grad_norm": 29.895591735839844,
|
| 146 |
+
"learning_rate": 0.0003,
|
| 147 |
+
"loss": 40.4991640625,
|
| 148 |
+
"step": 12000
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 1.6515276630883569,
|
| 152 |
+
"eval_accuracy": 0.7232843609541934,
|
| 153 |
+
"eval_loss": 9.78125,
|
| 154 |
+
"eval_runtime": 170.7512,
|
| 155 |
+
"eval_samples_per_second": 9682.91,
|
| 156 |
+
"eval_steps_per_second": 18.916,
|
| 157 |
+
"step": 12000
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 1.7891549683457197,
|
| 161 |
+
"grad_norm": 16.52672576904297,
|
| 162 |
+
"learning_rate": 0.0003,
|
| 163 |
+
"loss": 40.5724921875,
|
| 164 |
+
"step": 13000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.9267822736030829,
|
| 168 |
+
"grad_norm": 21.801998138427734,
|
| 169 |
+
"learning_rate": 0.0003,
|
| 170 |
+
"loss": 40.29026953125,
|
| 171 |
+
"step": 14000
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.9267822736030829,
|
| 175 |
+
"eval_accuracy": 0.7247579731972368,
|
| 176 |
+
"eval_loss": 9.7109375,
|
| 177 |
+
"eval_runtime": 169.376,
|
| 178 |
+
"eval_samples_per_second": 9761.53,
|
| 179 |
+
"eval_steps_per_second": 19.07,
|
| 180 |
+
"step": 14000
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 2.0644095788604457,
|
| 184 |
+
"grad_norm": 16.79710578918457,
|
| 185 |
+
"learning_rate": 0.0003,
|
| 186 |
+
"loss": 40.192109375,
|
| 187 |
+
"step": 15000
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 2.202036884117809,
|
| 191 |
+
"grad_norm": 19.166486740112305,
|
| 192 |
+
"learning_rate": 0.0003,
|
| 193 |
+
"loss": 39.6695,
|
| 194 |
+
"step": 16000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 2.202036884117809,
|
| 198 |
+
"eval_accuracy": 0.726963905479327,
|
| 199 |
+
"eval_loss": 9.6171875,
|
| 200 |
+
"eval_runtime": 170.9384,
|
| 201 |
+
"eval_samples_per_second": 9672.308,
|
| 202 |
+
"eval_steps_per_second": 18.896,
|
| 203 |
+
"step": 16000
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 2.339664189375172,
|
| 207 |
+
"grad_norm": 15.058335304260254,
|
| 208 |
+
"learning_rate": 0.0003,
|
| 209 |
+
"loss": 39.4829609375,
|
| 210 |
+
"step": 17000
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 2.477291494632535,
|
| 214 |
+
"grad_norm": 15.027190208435059,
|
| 215 |
+
"learning_rate": 0.0003,
|
| 216 |
+
"loss": 39.3190859375,
|
| 217 |
+
"step": 18000
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 2.477291494632535,
|
| 221 |
+
"eval_accuracy": 0.7287604009112041,
|
| 222 |
+
"eval_loss": 9.546875,
|
| 223 |
+
"eval_runtime": 170.8414,
|
| 224 |
+
"eval_samples_per_second": 9677.8,
|
| 225 |
+
"eval_steps_per_second": 18.906,
|
| 226 |
+
"step": 18000
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 2.614918799889898,
|
| 230 |
+
"grad_norm": 17.88246726989746,
|
| 231 |
+
"learning_rate": 0.0003,
|
| 232 |
+
"loss": 39.1439296875,
|
| 233 |
+
"step": 19000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 2.7525461051472613,
|
| 237 |
+
"grad_norm": 63.608863830566406,
|
| 238 |
+
"learning_rate": 0.0003,
|
| 239 |
+
"loss": 39.0614765625,
|
| 240 |
+
"step": 20000
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 2.7525461051472613,
|
| 244 |
+
"eval_accuracy": 0.7304020938141926,
|
| 245 |
+
"eval_loss": 9.46875,
|
| 246 |
+
"eval_runtime": 169.4479,
|
| 247 |
+
"eval_samples_per_second": 9757.391,
|
| 248 |
+
"eval_steps_per_second": 19.062,
|
| 249 |
+
"step": 20000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.8901734104046244,
|
| 253 |
+
"grad_norm": 16.970745086669922,
|
| 254 |
+
"learning_rate": 0.0003,
|
| 255 |
+
"loss": 38.933671875,
|
| 256 |
+
"step": 21000
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 3.0278007156619875,
|
| 260 |
+
"grad_norm": 19.159212112426758,
|
| 261 |
+
"learning_rate": 0.0003,
|
| 262 |
+
"loss": 39.2828828125,
|
| 263 |
+
"step": 22000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 3.0278007156619875,
|
| 267 |
+
"eval_accuracy": 0.7285464859640399,
|
| 268 |
+
"eval_loss": 9.5390625,
|
| 269 |
+
"eval_runtime": 168.2153,
|
| 270 |
+
"eval_samples_per_second": 9828.886,
|
| 271 |
+
"eval_steps_per_second": 19.202,
|
| 272 |
+
"step": 22000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 3.1654280209193506,
|
| 276 |
+
"grad_norm": 28.30343246459961,
|
| 277 |
+
"learning_rate": 0.0003,
|
| 278 |
+
"loss": 39.1001796875,
|
| 279 |
+
"step": 23000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 3.3030553261767133,
|
| 283 |
+
"grad_norm": 14.642151832580566,
|
| 284 |
+
"learning_rate": 0.0003,
|
| 285 |
+
"loss": 38.792859375,
|
| 286 |
+
"step": 24000
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 3.3030553261767133,
|
| 290 |
+
"eval_accuracy": 0.7310024378583697,
|
| 291 |
+
"eval_loss": 9.421875,
|
| 292 |
+
"eval_runtime": 168.799,
|
| 293 |
+
"eval_samples_per_second": 9794.899,
|
| 294 |
+
"eval_steps_per_second": 19.135,
|
| 295 |
+
"step": 24000
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 3.4406826314340764,
|
| 299 |
+
"grad_norm": 32.9803352355957,
|
| 300 |
+
"learning_rate": 0.0003,
|
| 301 |
+
"loss": 38.72437890625,
|
| 302 |
+
"step": 25000
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 3.5783099366914395,
|
| 306 |
+
"grad_norm": 15.326594352722168,
|
| 307 |
+
"learning_rate": 0.0003,
|
| 308 |
+
"loss": 38.7340703125,
|
| 309 |
+
"step": 26000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 3.5783099366914395,
|
| 313 |
+
"eval_accuracy": 0.7314286407398777,
|
| 314 |
+
"eval_loss": 9.390625,
|
| 315 |
+
"eval_runtime": 168.4702,
|
| 316 |
+
"eval_samples_per_second": 9814.015,
|
| 317 |
+
"eval_steps_per_second": 19.173,
|
| 318 |
+
"step": 26000
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 3.7159372419488026,
|
| 322 |
+
"grad_norm": 17.456626892089844,
|
| 323 |
+
"learning_rate": 0.0003,
|
| 324 |
+
"loss": 38.50905078125,
|
| 325 |
+
"step": 27000
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 3.8535645472061657,
|
| 329 |
+
"grad_norm": 19.74641990661621,
|
| 330 |
+
"learning_rate": 0.0003,
|
| 331 |
+
"loss": 38.4493125,
|
| 332 |
+
"step": 28000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 3.8535645472061657,
|
| 336 |
+
"eval_accuracy": 0.7326063700018453,
|
| 337 |
+
"eval_loss": 9.3359375,
|
| 338 |
+
"eval_runtime": 167.8973,
|
| 339 |
+
"eval_samples_per_second": 9847.504,
|
| 340 |
+
"eval_steps_per_second": 19.238,
|
| 341 |
+
"step": 28000
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 3.991191852463529,
|
| 345 |
+
"grad_norm": 16.116609573364258,
|
| 346 |
+
"learning_rate": 0.0003,
|
| 347 |
+
"loss": 38.4295859375,
|
| 348 |
+
"step": 29000
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 4.1288191577208915,
|
| 352 |
+
"grad_norm": 14.791277885437012,
|
| 353 |
+
"learning_rate": 0.0003,
|
| 354 |
+
"loss": 38.2604921875,
|
| 355 |
+
"step": 30000
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 4.1288191577208915,
|
| 359 |
+
"eval_accuracy": 0.73262638229311,
|
| 360 |
+
"eval_loss": 9.34375,
|
| 361 |
+
"eval_runtime": 169.3505,
|
| 362 |
+
"eval_samples_per_second": 9763.003,
|
| 363 |
+
"eval_steps_per_second": 19.073,
|
| 364 |
+
"step": 30000
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 4.266446462978255,
|
| 368 |
+
"grad_norm": 14.701274871826172,
|
| 369 |
+
"learning_rate": 0.0003,
|
| 370 |
+
"loss": 38.178046875,
|
| 371 |
+
"step": 31000
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 4.404073768235618,
|
| 375 |
+
"grad_norm": 32.28364181518555,
|
| 376 |
+
"learning_rate": 0.0003,
|
| 377 |
+
"loss": 38.118671875,
|
| 378 |
+
"step": 32000
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 4.404073768235618,
|
| 382 |
+
"eval_accuracy": 0.7330061560700385,
|
| 383 |
+
"eval_loss": 9.328125,
|
| 384 |
+
"eval_runtime": 169.1911,
|
| 385 |
+
"eval_samples_per_second": 9772.196,
|
| 386 |
+
"eval_steps_per_second": 19.091,
|
| 387 |
+
"step": 32000
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 4.541701073492981,
|
| 391 |
+
"grad_norm": 14.65267562866211,
|
| 392 |
+
"learning_rate": 0.0003,
|
| 393 |
+
"loss": 38.04671484375,
|
| 394 |
+
"step": 33000
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 4.679328378750344,
|
| 398 |
+
"grad_norm": 15.814043045043945,
|
| 399 |
+
"learning_rate": 0.0003,
|
| 400 |
+
"loss": 38.03919140625,
|
| 401 |
+
"step": 34000
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 4.679328378750344,
|
| 405 |
+
"eval_accuracy": 0.7350299683164979,
|
| 406 |
+
"eval_loss": 9.234375,
|
| 407 |
+
"eval_runtime": 170.3006,
|
| 408 |
+
"eval_samples_per_second": 9708.532,
|
| 409 |
+
"eval_steps_per_second": 18.966,
|
| 410 |
+
"step": 34000
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 4.8169556840077075,
|
| 414 |
+
"grad_norm": 22.082040786743164,
|
| 415 |
+
"learning_rate": 0.0003,
|
| 416 |
+
"loss": 38.00453125,
|
| 417 |
+
"step": 35000
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 4.95458298926507,
|
| 421 |
+
"grad_norm": 20.287931442260742,
|
| 422 |
+
"learning_rate": 0.0003,
|
| 423 |
+
"loss": 37.977375,
|
| 424 |
+
"step": 36000
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 4.95458298926507,
|
| 428 |
+
"eval_accuracy": 0.7339120258458323,
|
| 429 |
+
"eval_loss": 9.28125,
|
| 430 |
+
"eval_runtime": 171.801,
|
| 431 |
+
"eval_samples_per_second": 9623.744,
|
| 432 |
+
"eval_steps_per_second": 18.801,
|
| 433 |
+
"step": 36000
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 5.092210294522434,
|
| 437 |
+
"grad_norm": 30.58030128479004,
|
| 438 |
+
"learning_rate": 0.0003,
|
| 439 |
+
"loss": 37.9320703125,
|
| 440 |
+
"step": 37000
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 5.229837599779796,
|
| 444 |
+
"grad_norm": 19.570669174194336,
|
| 445 |
+
"learning_rate": 0.0003,
|
| 446 |
+
"loss": 37.9665234375,
|
| 447 |
+
"step": 38000
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 5.229837599779796,
|
| 451 |
+
"eval_accuracy": 0.7340658881695064,
|
| 452 |
+
"eval_loss": 9.28125,
|
| 453 |
+
"eval_runtime": 167.7847,
|
| 454 |
+
"eval_samples_per_second": 9854.11,
|
| 455 |
+
"eval_steps_per_second": 19.251,
|
| 456 |
+
"step": 38000
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 5.367464905037159,
|
| 460 |
+
"grad_norm": 17.29003143310547,
|
| 461 |
+
"learning_rate": 0.0003,
|
| 462 |
+
"loss": 37.9463984375,
|
| 463 |
+
"step": 39000
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 5.505092210294523,
|
| 467 |
+
"grad_norm": 16.568086624145508,
|
| 468 |
+
"learning_rate": 0.0003,
|
| 469 |
+
"loss": 37.7375625,
|
| 470 |
+
"step": 40000
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 5.505092210294523,
|
| 474 |
+
"eval_accuracy": 0.735108947509515,
|
| 475 |
+
"eval_loss": 9.2265625,
|
| 476 |
+
"eval_runtime": 170.3335,
|
| 477 |
+
"eval_samples_per_second": 9706.659,
|
| 478 |
+
"eval_steps_per_second": 18.963,
|
| 479 |
+
"step": 40000
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 5.642719515551885,
|
| 483 |
+
"grad_norm": 110.49578094482422,
|
| 484 |
+
"learning_rate": 0.0003,
|
| 485 |
+
"loss": 37.6945,
|
| 486 |
+
"step": 41000
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 5.780346820809249,
|
| 490 |
+
"grad_norm": 24.624027252197266,
|
| 491 |
+
"learning_rate": 0.0003,
|
| 492 |
+
"loss": 37.72793359375,
|
| 493 |
+
"step": 42000
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 5.780346820809249,
|
| 497 |
+
"eval_accuracy": 0.7357611562920721,
|
| 498 |
+
"eval_loss": 9.203125,
|
| 499 |
+
"eval_runtime": 168.0213,
|
| 500 |
+
"eval_samples_per_second": 9840.236,
|
| 501 |
+
"eval_steps_per_second": 19.224,
|
| 502 |
+
"step": 42000
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 5.917974126066611,
|
| 506 |
+
"grad_norm": 17.721742630004883,
|
| 507 |
+
"learning_rate": 0.0003,
|
| 508 |
+
"loss": 37.635171875,
|
| 509 |
+
"step": 43000
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 6.055601431323975,
|
| 513 |
+
"grad_norm": 14.108975410461426,
|
| 514 |
+
"learning_rate": 0.0003,
|
| 515 |
+
"loss": 37.572765625,
|
| 516 |
+
"step": 44000
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 6.055601431323975,
|
| 520 |
+
"eval_accuracy": 0.7358946142236276,
|
| 521 |
+
"eval_loss": 9.1953125,
|
| 522 |
+
"eval_runtime": 168.7352,
|
| 523 |
+
"eval_samples_per_second": 9798.601,
|
| 524 |
+
"eval_steps_per_second": 19.142,
|
| 525 |
+
"step": 44000
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 6.193228736581338,
|
| 529 |
+
"grad_norm": 22.010786056518555,
|
| 530 |
+
"learning_rate": 0.0003,
|
| 531 |
+
"loss": 37.4784453125,
|
| 532 |
+
"step": 45000
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 6.330856041838701,
|
| 536 |
+
"grad_norm": 28.03326988220215,
|
| 537 |
+
"learning_rate": 0.0003,
|
| 538 |
+
"loss": 37.5036328125,
|
| 539 |
+
"step": 46000
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 6.330856041838701,
|
| 543 |
+
"eval_accuracy": 0.7368965623776919,
|
| 544 |
+
"eval_loss": 9.171875,
|
| 545 |
+
"eval_runtime": 166.8883,
|
| 546 |
+
"eval_samples_per_second": 9907.04,
|
| 547 |
+
"eval_steps_per_second": 19.354,
|
| 548 |
+
"step": 46000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 6.468483347096064,
|
| 552 |
+
"grad_norm": 17.36493492126465,
|
| 553 |
+
"learning_rate": 0.0003,
|
| 554 |
+
"loss": 37.4494453125,
|
| 555 |
+
"step": 47000
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 6.6061106523534265,
|
| 559 |
+
"grad_norm": 19.51688003540039,
|
| 560 |
+
"learning_rate": 0.0003,
|
| 561 |
+
"loss": 37.435234375,
|
| 562 |
+
"step": 48000
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 6.6061106523534265,
|
| 566 |
+
"eval_accuracy": 0.7361287527698055,
|
| 567 |
+
"eval_loss": 9.1796875,
|
| 568 |
+
"eval_runtime": 169.5516,
|
| 569 |
+
"eval_samples_per_second": 9751.421,
|
| 570 |
+
"eval_steps_per_second": 19.05,
|
| 571 |
+
"step": 48000
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 6.74373795761079,
|
| 575 |
+
"grad_norm": 15.679317474365234,
|
| 576 |
+
"learning_rate": 0.0003,
|
| 577 |
+
"loss": 37.3981484375,
|
| 578 |
+
"step": 49000
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 6.881365262868153,
|
| 582 |
+
"grad_norm": 42.613555908203125,
|
| 583 |
+
"learning_rate": 0.0003,
|
| 584 |
+
"loss": 37.51515625,
|
| 585 |
+
"step": 50000
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 6.881365262868153,
|
| 589 |
+
"eval_accuracy": 0.7378156733015306,
|
| 590 |
+
"eval_loss": 9.125,
|
| 591 |
+
"eval_runtime": 170.8004,
|
| 592 |
+
"eval_samples_per_second": 9680.126,
|
| 593 |
+
"eval_steps_per_second": 18.911,
|
| 594 |
+
"step": 50000
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 7.018992568125516,
|
| 598 |
+
"grad_norm": 15.565272331237793,
|
| 599 |
+
"learning_rate": 0.0003,
|
| 600 |
+
"loss": 37.4844375,
|
| 601 |
+
"step": 51000
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 7.156619873382879,
|
| 605 |
+
"grad_norm": 15.575469017028809,
|
| 606 |
+
"learning_rate": 0.0003,
|
| 607 |
+
"loss": 37.2694375,
|
| 608 |
+
"step": 52000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 7.156619873382879,
|
| 612 |
+
"eval_accuracy": 0.7383814207101249,
|
| 613 |
+
"eval_loss": 9.09375,
|
| 614 |
+
"eval_runtime": 167.1627,
|
| 615 |
+
"eval_samples_per_second": 9890.776,
|
| 616 |
+
"eval_steps_per_second": 19.322,
|
| 617 |
+
"step": 52000
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 7.2942471786402425,
|
| 621 |
+
"grad_norm": 15.146500587463379,
|
| 622 |
+
"learning_rate": 0.0003,
|
| 623 |
+
"loss": 37.243421875,
|
| 624 |
+
"step": 53000
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 7.431874483897605,
|
| 628 |
+
"grad_norm": 17.28059959411621,
|
| 629 |
+
"learning_rate": 0.0003,
|
| 630 |
+
"loss": 37.23215625,
|
| 631 |
+
"step": 54000
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 7.431874483897605,
|
| 635 |
+
"eval_accuracy": 0.7379147365824262,
|
| 636 |
+
"eval_loss": 9.1015625,
|
| 637 |
+
"eval_runtime": 169.7822,
|
| 638 |
+
"eval_samples_per_second": 9738.179,
|
| 639 |
+
"eval_steps_per_second": 19.024,
|
| 640 |
+
"step": 54000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 7.569501789154968,
|
| 644 |
+
"grad_norm": 29.069555282592773,
|
| 645 |
+
"learning_rate": 0.0003,
|
| 646 |
+
"loss": 37.22365625,
|
| 647 |
+
"step": 55000
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 7.707129094412331,
|
| 651 |
+
"grad_norm": 26.9329833984375,
|
| 652 |
+
"learning_rate": 0.0003,
|
| 653 |
+
"loss": 37.22718359375,
|
| 654 |
+
"step": 56000
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 7.707129094412331,
|
| 658 |
+
"eval_accuracy": 0.7364887765974263,
|
| 659 |
+
"eval_loss": 9.171875,
|
| 660 |
+
"eval_runtime": 170.0511,
|
| 661 |
+
"eval_samples_per_second": 9722.778,
|
| 662 |
+
"eval_steps_per_second": 18.994,
|
| 663 |
+
"step": 56000
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 7.844756399669695,
|
| 667 |
+
"grad_norm": 97.25556182861328,
|
| 668 |
+
"learning_rate": 0.0003,
|
| 669 |
+
"loss": 40.3630625,
|
| 670 |
+
"step": 57000
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 7.982383704927058,
|
| 674 |
+
"grad_norm": 19.714794158935547,
|
| 675 |
+
"learning_rate": 0.0003,
|
| 676 |
+
"loss": 37.60580859375,
|
| 677 |
+
"step": 58000
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 7.982383704927058,
|
| 681 |
+
"eval_accuracy": 0.7352168040424968,
|
| 682 |
+
"eval_loss": 9.234375,
|
| 683 |
+
"eval_runtime": 169.0419,
|
| 684 |
+
"eval_samples_per_second": 9780.822,
|
| 685 |
+
"eval_steps_per_second": 19.108,
|
| 686 |
+
"step": 58000
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 8.12001101018442,
|
| 690 |
+
"grad_norm": 14.792304039001465,
|
| 691 |
+
"learning_rate": 0.0003,
|
| 692 |
+
"loss": 37.54733984375,
|
| 693 |
+
"step": 59000
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 8.257638315441783,
|
| 697 |
+
"grad_norm": 15.469705581665039,
|
| 698 |
+
"learning_rate": 0.0003,
|
| 699 |
+
"loss": 37.5459453125,
|
| 700 |
+
"step": 60000
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 8.257638315441783,
|
| 704 |
+
"eval_accuracy": 0.7371664310551599,
|
| 705 |
+
"eval_loss": 9.1328125,
|
| 706 |
+
"eval_runtime": 166.3796,
|
| 707 |
+
"eval_samples_per_second": 9937.33,
|
| 708 |
+
"eval_steps_per_second": 19.413,
|
| 709 |
+
"step": 60000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 8.395265620699147,
|
| 713 |
+
"grad_norm": 15.161516189575195,
|
| 714 |
+
"learning_rate": 0.0003,
|
| 715 |
+
"loss": 37.47540625,
|
| 716 |
+
"step": 61000
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 8.53289292595651,
|
| 720 |
+
"grad_norm": 14.802396774291992,
|
| 721 |
+
"learning_rate": 0.0003,
|
| 722 |
+
"loss": 37.30961328125,
|
| 723 |
+
"step": 62000
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 8.53289292595651,
|
| 727 |
+
"eval_accuracy": 0.7381438951141913,
|
| 728 |
+
"eval_loss": 9.09375,
|
| 729 |
+
"eval_runtime": 168.1235,
|
| 730 |
+
"eval_samples_per_second": 9834.251,
|
| 731 |
+
"eval_steps_per_second": 19.212,
|
| 732 |
+
"step": 62000
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 8.670520231213873,
|
| 736 |
+
"grad_norm": 20.38652229309082,
|
| 737 |
+
"learning_rate": 0.0003,
|
| 738 |
+
"loss": 37.20555859375,
|
| 739 |
+
"step": 63000
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 8.808147536471235,
|
| 743 |
+
"grad_norm": 15.676960945129395,
|
| 744 |
+
"learning_rate": 0.0003,
|
| 745 |
+
"loss": 37.14574609375,
|
| 746 |
+
"step": 64000
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 8.808147536471235,
|
| 750 |
+
"eval_accuracy": 0.7382565867319608,
|
| 751 |
+
"eval_loss": 9.1015625,
|
| 752 |
+
"eval_runtime": 171.8616,
|
| 753 |
+
"eval_samples_per_second": 9620.35,
|
| 754 |
+
"eval_steps_per_second": 18.794,
|
| 755 |
+
"step": 64000
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 8.9457748417286,
|
| 759 |
+
"grad_norm": 16.668174743652344,
|
| 760 |
+
"learning_rate": 0.0003,
|
| 761 |
+
"loss": 37.125671875,
|
| 762 |
+
"step": 65000
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 9.083402146985962,
|
| 766 |
+
"grad_norm": 16.503215789794922,
|
| 767 |
+
"learning_rate": 0.0003,
|
| 768 |
+
"loss": 37.01801171875,
|
| 769 |
+
"step": 66000
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 9.083402146985962,
|
| 773 |
+
"eval_accuracy": 0.7393124677323514,
|
| 774 |
+
"eval_loss": 9.046875,
|
| 775 |
+
"eval_runtime": 168.277,
|
| 776 |
+
"eval_samples_per_second": 9825.282,
|
| 777 |
+
"eval_steps_per_second": 19.195,
|
| 778 |
+
"step": 66000
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 9.221029452243325,
|
| 782 |
+
"grad_norm": 19.62793731689453,
|
| 783 |
+
"learning_rate": 0.0003,
|
| 784 |
+
"loss": 36.9910703125,
|
| 785 |
+
"step": 67000
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 9.358656757500688,
|
| 789 |
+
"grad_norm": 18.604116439819336,
|
| 790 |
+
"learning_rate": 0.0003,
|
| 791 |
+
"loss": 36.9631328125,
|
| 792 |
+
"step": 68000
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 9.358656757500688,
|
| 796 |
+
"eval_accuracy": 0.7395416473430355,
|
| 797 |
+
"eval_loss": 9.03125,
|
| 798 |
+
"eval_runtime": 166.5578,
|
| 799 |
+
"eval_samples_per_second": 9926.697,
|
| 800 |
+
"eval_steps_per_second": 19.393,
|
| 801 |
+
"step": 68000
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 9.49628406275805,
|
| 805 |
+
"grad_norm": 20.98838233947754,
|
| 806 |
+
"learning_rate": 0.0003,
|
| 807 |
+
"loss": 36.963703125,
|
| 808 |
+
"step": 69000
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 9.633911368015415,
|
| 812 |
+
"grad_norm": 17.784839630126953,
|
| 813 |
+
"learning_rate": 0.0003,
|
| 814 |
+
"loss": 36.96887109375,
|
| 815 |
+
"step": 70000
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 9.633911368015415,
|
| 819 |
+
"eval_accuracy": 0.7390765595619299,
|
| 820 |
+
"eval_loss": 9.0625,
|
| 821 |
+
"eval_runtime": 166.5724,
|
| 822 |
+
"eval_samples_per_second": 9925.827,
|
| 823 |
+
"eval_steps_per_second": 19.391,
|
| 824 |
+
"step": 70000
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 9.771538673272778,
|
| 828 |
+
"grad_norm": 27.863100051879883,
|
| 829 |
+
"learning_rate": 0.0003,
|
| 830 |
+
"loss": 37.0048671875,
|
| 831 |
+
"step": 71000
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 9.90916597853014,
|
| 835 |
+
"grad_norm": 16.188459396362305,
|
| 836 |
+
"learning_rate": 0.0003,
|
| 837 |
+
"loss": 36.9179296875,
|
| 838 |
+
"step": 72000
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 9.90916597853014,
|
| 842 |
+
"eval_accuracy": 0.7394581492526459,
|
| 843 |
+
"eval_loss": 9.03125,
|
| 844 |
+
"eval_runtime": 170.082,
|
| 845 |
+
"eval_samples_per_second": 9721.013,
|
| 846 |
+
"eval_steps_per_second": 18.991,
|
| 847 |
+
"step": 72000
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 10.046793283787503,
|
| 851 |
+
"grad_norm": 14.628756523132324,
|
| 852 |
+
"learning_rate": 0.0003,
|
| 853 |
+
"loss": 36.888484375,
|
| 854 |
+
"step": 73000
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 10.184420589044867,
|
| 858 |
+
"grad_norm": 14.934839248657227,
|
| 859 |
+
"learning_rate": 0.0003,
|
| 860 |
+
"loss": 36.7855625,
|
| 861 |
+
"step": 74000
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 10.184420589044867,
|
| 865 |
+
"eval_accuracy": 0.7404586215139355,
|
| 866 |
+
"eval_loss": 8.9921875,
|
| 867 |
+
"eval_runtime": 168.2373,
|
| 868 |
+
"eval_samples_per_second": 9827.6,
|
| 869 |
+
"eval_steps_per_second": 19.199,
|
| 870 |
+
"step": 74000
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 10.32204789430223,
|
| 874 |
+
"grad_norm": 15.220124244689941,
|
| 875 |
+
"learning_rate": 0.0003,
|
| 876 |
+
"loss": 36.9548515625,
|
| 877 |
+
"step": 75000
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 10.459675199559593,
|
| 881 |
+
"grad_norm": 15.933542251586914,
|
| 882 |
+
"learning_rate": 0.0003,
|
| 883 |
+
"loss": 36.85228125,
|
| 884 |
+
"step": 76000
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 10.459675199559593,
|
| 888 |
+
"eval_accuracy": 0.7398413045299678,
|
| 889 |
+
"eval_loss": 9.0234375,
|
| 890 |
+
"eval_runtime": 169.5447,
|
| 891 |
+
"eval_samples_per_second": 9751.82,
|
| 892 |
+
"eval_steps_per_second": 19.051,
|
| 893 |
+
"step": 76000
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 10.597302504816955,
|
| 897 |
+
"grad_norm": 15.21678638458252,
|
| 898 |
+
"learning_rate": 0.0003,
|
| 899 |
+
"loss": 36.8666171875,
|
| 900 |
+
"step": 77000
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 10.734929810074318,
|
| 904 |
+
"grad_norm": 16.906696319580078,
|
| 905 |
+
"learning_rate": 0.0003,
|
| 906 |
+
"loss": 36.78946875,
|
| 907 |
+
"step": 78000
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 10.734929810074318,
|
| 911 |
+
"eval_accuracy": 0.7396672771453636,
|
| 912 |
+
"eval_loss": 9.015625,
|
| 913 |
+
"eval_runtime": 168.0605,
|
| 914 |
+
"eval_samples_per_second": 9837.938,
|
| 915 |
+
"eval_steps_per_second": 19.219,
|
| 916 |
+
"step": 78000
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 10.872557115331682,
|
| 920 |
+
"grad_norm": 15.474593162536621,
|
| 921 |
+
"learning_rate": 0.0003,
|
| 922 |
+
"loss": 36.7689296875,
|
| 923 |
+
"step": 79000
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 11.010184420589045,
|
| 927 |
+
"grad_norm": 14.83968448638916,
|
| 928 |
+
"learning_rate": 0.0003,
|
| 929 |
+
"loss": 36.761296875,
|
| 930 |
+
"step": 80000
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 11.010184420589045,
|
| 934 |
+
"eval_accuracy": 0.7401702144442798,
|
| 935 |
+
"eval_loss": 9.0,
|
| 936 |
+
"eval_runtime": 167.1731,
|
| 937 |
+
"eval_samples_per_second": 9890.16,
|
| 938 |
+
"eval_steps_per_second": 19.321,
|
| 939 |
+
"step": 80000
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 11.147811725846408,
|
| 943 |
+
"grad_norm": 16.945621490478516,
|
| 944 |
+
"learning_rate": 0.0003,
|
| 945 |
+
"loss": 36.7302734375,
|
| 946 |
+
"step": 81000
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 11.28543903110377,
|
| 950 |
+
"grad_norm": 20.112407684326172,
|
| 951 |
+
"learning_rate": 0.0003,
|
| 952 |
+
"loss": 36.74803125,
|
| 953 |
+
"step": 82000
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 11.28543903110377,
|
| 957 |
+
"eval_accuracy": 0.7397283390276018,
|
| 958 |
+
"eval_loss": 9.046875,
|
| 959 |
+
"eval_runtime": 167.1017,
|
| 960 |
+
"eval_samples_per_second": 9894.387,
|
| 961 |
+
"eval_steps_per_second": 19.33,
|
| 962 |
+
"step": 82000
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 11.423066336361135,
|
| 966 |
+
"grad_norm": 15.264861106872559,
|
| 967 |
+
"learning_rate": 0.0003,
|
| 968 |
+
"loss": 36.7436875,
|
| 969 |
+
"step": 83000
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 11.560693641618498,
|
| 973 |
+
"grad_norm": 35.09538269042969,
|
| 974 |
+
"learning_rate": 0.0003,
|
| 975 |
+
"loss": 36.68840625,
|
| 976 |
+
"step": 84000
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 11.560693641618498,
|
| 980 |
+
"eval_accuracy": 0.7408698045825953,
|
| 981 |
+
"eval_loss": 8.96875,
|
| 982 |
+
"eval_runtime": 168.7302,
|
| 983 |
+
"eval_samples_per_second": 9798.89,
|
| 984 |
+
"eval_steps_per_second": 19.143,
|
| 985 |
+
"step": 84000
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 11.69832094687586,
|
| 989 |
+
"grad_norm": 19.97286605834961,
|
| 990 |
+
"learning_rate": 0.0003,
|
| 991 |
+
"loss": 36.69001171875,
|
| 992 |
+
"step": 85000
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 11.835948252133223,
|
| 996 |
+
"grad_norm": 17.521934509277344,
|
| 997 |
+
"learning_rate": 0.0003,
|
| 998 |
+
"loss": 36.67117578125,
|
| 999 |
+
"step": 86000
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 11.835948252133223,
|
| 1003 |
+
"eval_accuracy": 0.7407812017839025,
|
| 1004 |
+
"eval_loss": 8.9765625,
|
| 1005 |
+
"eval_runtime": 169.2449,
|
| 1006 |
+
"eval_samples_per_second": 9769.092,
|
| 1007 |
+
"eval_steps_per_second": 19.085,
|
| 1008 |
+
"step": 86000
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 11.973575557390586,
|
| 1012 |
+
"grad_norm": 15.003190040588379,
|
| 1013 |
+
"learning_rate": 0.0003,
|
| 1014 |
+
"loss": 36.64848046875,
|
| 1015 |
+
"step": 87000
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 12.11120286264795,
|
| 1019 |
+
"grad_norm": 20.016794204711914,
|
| 1020 |
+
"learning_rate": 0.0003,
|
| 1021 |
+
"loss": 36.72609765625,
|
| 1022 |
+
"step": 88000
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 12.11120286264795,
|
| 1026 |
+
"eval_accuracy": 0.7402567288458476,
|
| 1027 |
+
"eval_loss": 9.0,
|
| 1028 |
+
"eval_runtime": 168.1699,
|
| 1029 |
+
"eval_samples_per_second": 9831.539,
|
| 1030 |
+
"eval_steps_per_second": 19.207,
|
| 1031 |
+
"step": 88000
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 12.248830167905313,
|
| 1035 |
+
"grad_norm": 16.020404815673828,
|
| 1036 |
+
"learning_rate": 0.0003,
|
| 1037 |
+
"loss": 36.8169140625,
|
| 1038 |
+
"step": 89000
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 12.386457473162675,
|
| 1042 |
+
"grad_norm": 16.005430221557617,
|
| 1043 |
+
"learning_rate": 0.0003,
|
| 1044 |
+
"loss": 36.652015625,
|
| 1045 |
+
"step": 90000
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 12.386457473162675,
|
| 1049 |
+
"eval_accuracy": 0.740178315482875,
|
| 1050 |
+
"eval_loss": 9.0,
|
| 1051 |
+
"eval_runtime": 169.3304,
|
| 1052 |
+
"eval_samples_per_second": 9764.16,
|
| 1053 |
+
"eval_steps_per_second": 19.075,
|
| 1054 |
+
"step": 90000
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 12.524084778420038,
|
| 1058 |
+
"grad_norm": 14.788511276245117,
|
| 1059 |
+
"learning_rate": 0.0003,
|
| 1060 |
+
"loss": 36.70846875,
|
| 1061 |
+
"step": 91000
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 12.661712083677402,
|
| 1065 |
+
"grad_norm": 15.171088218688965,
|
| 1066 |
+
"learning_rate": 0.0003,
|
| 1067 |
+
"loss": 36.9015078125,
|
| 1068 |
+
"step": 92000
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 12.661712083677402,
|
| 1072 |
+
"eval_accuracy": 0.7409616192832641,
|
| 1073 |
+
"eval_loss": 8.9765625,
|
| 1074 |
+
"eval_runtime": 167.4567,
|
| 1075 |
+
"eval_samples_per_second": 9873.414,
|
| 1076 |
+
"eval_steps_per_second": 19.289,
|
| 1077 |
+
"step": 92000
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 12.799339388934765,
|
| 1081 |
+
"grad_norm": 15.66518497467041,
|
| 1082 |
+
"learning_rate": 0.0003,
|
| 1083 |
+
"loss": 36.66136328125,
|
| 1084 |
+
"step": 93000
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 12.936966694192128,
|
| 1088 |
+
"grad_norm": 17.391189575195312,
|
| 1089 |
+
"learning_rate": 0.0003,
|
| 1090 |
+
"loss": 36.62637109375,
|
| 1091 |
+
"step": 94000
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 12.936966694192128,
|
| 1095 |
+
"eval_accuracy": 0.7408841425444469,
|
| 1096 |
+
"eval_loss": 8.96875,
|
| 1097 |
+
"eval_runtime": 168.9204,
|
| 1098 |
+
"eval_samples_per_second": 9787.86,
|
| 1099 |
+
"eval_steps_per_second": 19.121,
|
| 1100 |
+
"step": 94000
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 13.07459399944949,
|
| 1104 |
+
"grad_norm": 14.819647789001465,
|
| 1105 |
+
"learning_rate": 0.0003,
|
| 1106 |
+
"loss": 36.554375,
|
| 1107 |
+
"step": 95000
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 13.212221304706853,
|
| 1111 |
+
"grad_norm": 15.322741508483887,
|
| 1112 |
+
"learning_rate": 0.0003,
|
| 1113 |
+
"loss": 36.5329375,
|
| 1114 |
+
"step": 96000
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 13.212221304706853,
|
| 1118 |
+
"eval_accuracy": 0.7418048186655257,
|
| 1119 |
+
"eval_loss": 8.9296875,
|
| 1120 |
+
"eval_runtime": 168.0129,
|
| 1121 |
+
"eval_samples_per_second": 9840.727,
|
| 1122 |
+
"eval_steps_per_second": 19.225,
|
| 1123 |
+
"step": 96000
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 13.349848609964218,
|
| 1127 |
+
"grad_norm": 15.646768569946289,
|
| 1128 |
+
"learning_rate": 0.0003,
|
| 1129 |
+
"loss": 36.5148125,
|
| 1130 |
+
"step": 97000
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 13.48747591522158,
|
| 1134 |
+
"grad_norm": 22.885791778564453,
|
| 1135 |
+
"learning_rate": 0.0003,
|
| 1136 |
+
"loss": 36.550625,
|
| 1137 |
+
"step": 98000
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 13.48747591522158,
|
| 1141 |
+
"eval_accuracy": 0.7403524509758528,
|
| 1142 |
+
"eval_loss": 9.0078125,
|
| 1143 |
+
"eval_runtime": 169.9032,
|
| 1144 |
+
"eval_samples_per_second": 9731.239,
|
| 1145 |
+
"eval_steps_per_second": 19.011,
|
| 1146 |
+
"step": 98000
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 13.625103220478943,
|
| 1150 |
+
"grad_norm": 23.617691040039062,
|
| 1151 |
+
"learning_rate": 0.0003,
|
| 1152 |
+
"loss": 36.64653515625,
|
| 1153 |
+
"step": 99000
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 13.762730525736306,
|
| 1157 |
+
"grad_norm": 16.434221267700195,
|
| 1158 |
+
"learning_rate": 0.0003,
|
| 1159 |
+
"loss": 36.5250390625,
|
| 1160 |
+
"step": 100000
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 13.762730525736306,
|
| 1164 |
+
"eval_accuracy": 0.741298813475031,
|
| 1165 |
+
"eval_loss": 8.9453125,
|
| 1166 |
+
"eval_runtime": 170.5631,
|
| 1167 |
+
"eval_samples_per_second": 9693.591,
|
| 1168 |
+
"eval_steps_per_second": 18.937,
|
| 1169 |
+
"step": 100000
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 13.90035783099367,
|
| 1173 |
+
"grad_norm": 15.552685737609863,
|
| 1174 |
+
"learning_rate": 0.0003,
|
| 1175 |
+
"loss": 36.50655078125,
|
| 1176 |
+
"step": 101000
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 14.037985136251033,
|
| 1180 |
+
"grad_norm": 15.682257652282715,
|
| 1181 |
+
"learning_rate": 0.0003,
|
| 1182 |
+
"loss": 36.52517578125,
|
| 1183 |
+
"step": 102000
|
| 1184 |
+
},
|
| 1185 |
+
{
|
| 1186 |
+
"epoch": 14.037985136251033,
|
| 1187 |
+
"eval_accuracy": 0.7417503019794978,
|
| 1188 |
+
"eval_loss": 8.9296875,
|
| 1189 |
+
"eval_runtime": 167.2274,
|
| 1190 |
+
"eval_samples_per_second": 9886.95,
|
| 1191 |
+
"eval_steps_per_second": 19.315,
|
| 1192 |
+
"step": 102000
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 14.175612441508395,
|
| 1196 |
+
"grad_norm": 16.582626342773438,
|
| 1197 |
+
"learning_rate": 0.0003,
|
| 1198 |
+
"loss": 36.4439453125,
|
| 1199 |
+
"step": 103000
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 14.313239746765758,
|
| 1203 |
+
"grad_norm": 15.228682518005371,
|
| 1204 |
+
"learning_rate": 0.0003,
|
| 1205 |
+
"loss": 36.7357578125,
|
| 1206 |
+
"step": 104000
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 14.313239746765758,
|
| 1210 |
+
"eval_accuracy": 0.7409231190118309,
|
| 1211 |
+
"eval_loss": 8.984375,
|
| 1212 |
+
"eval_runtime": 168.495,
|
| 1213 |
+
"eval_samples_per_second": 9812.569,
|
| 1214 |
+
"eval_steps_per_second": 19.17,
|
| 1215 |
+
"step": 104000
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 14.45086705202312,
|
| 1219 |
+
"grad_norm": 18.01194190979004,
|
| 1220 |
+
"learning_rate": 0.0003,
|
| 1221 |
+
"loss": 36.506953125,
|
| 1222 |
+
"step": 105000
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 14.588494357280485,
|
| 1226 |
+
"grad_norm": 16.423683166503906,
|
| 1227 |
+
"learning_rate": 0.0003,
|
| 1228 |
+
"loss": 36.4571796875,
|
| 1229 |
+
"step": 106000
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 14.588494357280485,
|
| 1233 |
+
"eval_accuracy": 0.7415011536669796,
|
| 1234 |
+
"eval_loss": 8.9453125,
|
| 1235 |
+
"eval_runtime": 170.7059,
|
| 1236 |
+
"eval_samples_per_second": 9685.482,
|
| 1237 |
+
"eval_steps_per_second": 18.921,
|
| 1238 |
+
"step": 106000
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 14.726121662537848,
|
| 1242 |
+
"grad_norm": 15.935735702514648,
|
| 1243 |
+
"learning_rate": 0.0003,
|
| 1244 |
+
"loss": 36.4399453125,
|
| 1245 |
+
"step": 107000
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 14.86374896779521,
|
| 1249 |
+
"grad_norm": 34.532562255859375,
|
| 1250 |
+
"learning_rate": 0.0003,
|
| 1251 |
+
"loss": 36.496796875,
|
| 1252 |
+
"step": 108000
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"epoch": 14.86374896779521,
|
| 1256 |
+
"eval_accuracy": 0.740452263067793,
|
| 1257 |
+
"eval_loss": 9.0,
|
| 1258 |
+
"eval_runtime": 169.5593,
|
| 1259 |
+
"eval_samples_per_second": 9750.98,
|
| 1260 |
+
"eval_steps_per_second": 19.049,
|
| 1261 |
+
"step": 108000
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 15.001376273052573,
|
| 1265 |
+
"grad_norm": 17.5130615234375,
|
| 1266 |
+
"learning_rate": 0.0003,
|
| 1267 |
+
"loss": 36.44953125,
|
| 1268 |
+
"step": 109000
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 15.139003578309937,
|
| 1272 |
+
"grad_norm": 17.350488662719727,
|
| 1273 |
+
"learning_rate": 0.0003,
|
| 1274 |
+
"loss": 36.5754375,
|
| 1275 |
+
"step": 110000
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 15.139003578309937,
|
| 1279 |
+
"eval_accuracy": 0.7407140399047041,
|
| 1280 |
+
"eval_loss": 8.984375,
|
| 1281 |
+
"eval_runtime": 169.3269,
|
| 1282 |
+
"eval_samples_per_second": 9764.36,
|
| 1283 |
+
"eval_steps_per_second": 19.076,
|
| 1284 |
+
"step": 110000
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 15.2766308835673,
|
| 1288 |
+
"grad_norm": 16.463069915771484,
|
| 1289 |
+
"learning_rate": 0.0003,
|
| 1290 |
+
"loss": 36.57256640625,
|
| 1291 |
+
"step": 111000
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 15.414258188824663,
|
| 1295 |
+
"grad_norm": 15.29598331451416,
|
| 1296 |
+
"learning_rate": 0.0003,
|
| 1297 |
+
"loss": 36.57964453125,
|
| 1298 |
+
"step": 112000
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 15.414258188824663,
|
| 1302 |
+
"eval_accuracy": 0.7415344808420624,
|
| 1303 |
+
"eval_loss": 8.9375,
|
| 1304 |
+
"eval_runtime": 169.2422,
|
| 1305 |
+
"eval_samples_per_second": 9769.248,
|
| 1306 |
+
"eval_steps_per_second": 19.085,
|
| 1307 |
+
"step": 112000
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 15.551885494082025,
|
| 1311 |
+
"grad_norm": 26.094396591186523,
|
| 1312 |
+
"learning_rate": 0.0003,
|
| 1313 |
+
"loss": 36.45933203125,
|
| 1314 |
+
"step": 113000
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 15.689512799339388,
|
| 1318 |
+
"grad_norm": 15.42809009552002,
|
| 1319 |
+
"learning_rate": 0.0003,
|
| 1320 |
+
"loss": 36.4115546875,
|
| 1321 |
+
"step": 114000
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 15.689512799339388,
|
| 1325 |
+
"eval_accuracy": 0.7421097237159107,
|
| 1326 |
+
"eval_loss": 8.921875,
|
| 1327 |
+
"eval_runtime": 168.9547,
|
| 1328 |
+
"eval_samples_per_second": 9785.869,
|
| 1329 |
+
"eval_steps_per_second": 19.118,
|
| 1330 |
+
"step": 114000
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 15.827140104596753,
|
| 1334 |
+
"grad_norm": 18.432769775390625,
|
| 1335 |
+
"learning_rate": 0.0003,
|
| 1336 |
+
"loss": 36.36869921875,
|
| 1337 |
+
"step": 115000
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 15.964767409854115,
|
| 1341 |
+
"grad_norm": 16.479217529296875,
|
| 1342 |
+
"learning_rate": 0.0003,
|
| 1343 |
+
"loss": 36.33540625,
|
| 1344 |
+
"step": 116000
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 15.964767409854115,
|
| 1348 |
+
"eval_accuracy": 0.742688496326886,
|
| 1349 |
+
"eval_loss": 8.9140625,
|
| 1350 |
+
"eval_runtime": 171.1574,
|
| 1351 |
+
"eval_samples_per_second": 9659.935,
|
| 1352 |
+
"eval_steps_per_second": 18.872,
|
| 1353 |
+
"step": 116000
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 16.10239471511148,
|
| 1357 |
+
"grad_norm": 16.778072357177734,
|
| 1358 |
+
"learning_rate": 0.0003,
|
| 1359 |
+
"loss": 36.3029140625,
|
| 1360 |
+
"step": 117000
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 16.24002202036884,
|
| 1364 |
+
"grad_norm": 17.59860610961914,
|
| 1365 |
+
"learning_rate": 0.0003,
|
| 1366 |
+
"loss": 36.37992578125,
|
| 1367 |
+
"step": 118000
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 16.24002202036884,
|
| 1371 |
+
"eval_accuracy": 0.7424896095870896,
|
| 1372 |
+
"eval_loss": 8.9140625,
|
| 1373 |
+
"eval_runtime": 169.4524,
|
| 1374 |
+
"eval_samples_per_second": 9757.132,
|
| 1375 |
+
"eval_steps_per_second": 19.061,
|
| 1376 |
+
"step": 118000
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 16.377649325626205,
|
| 1380 |
+
"grad_norm": 19.460819244384766,
|
| 1381 |
+
"learning_rate": 0.0003,
|
| 1382 |
+
"loss": 36.3964765625,
|
| 1383 |
+
"step": 119000
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 16.515276630883566,
|
| 1387 |
+
"grad_norm": 47.38739776611328,
|
| 1388 |
+
"learning_rate": 0.0003,
|
| 1389 |
+
"loss": 36.399671875,
|
| 1390 |
+
"step": 120000
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 16.515276630883566,
|
| 1394 |
+
"eval_accuracy": 0.7421772826656658,
|
| 1395 |
+
"eval_loss": 8.90625,
|
| 1396 |
+
"eval_runtime": 168.2033,
|
| 1397 |
+
"eval_samples_per_second": 9829.585,
|
| 1398 |
+
"eval_steps_per_second": 19.203,
|
| 1399 |
+
"step": 120000
|
| 1400 |
+
},
|
| 1401 |
+
{
|
| 1402 |
+
"epoch": 16.515276630883566,
|
| 1403 |
+
"step": 120000,
|
| 1404 |
+
"total_flos": 1.617130434771026e+19,
|
| 1405 |
+
"train_loss": 38.9186564453125,
|
| 1406 |
+
"train_runtime": 44793.9304,
|
| 1407 |
+
"train_samples_per_second": 5486.458,
|
| 1408 |
+
"train_steps_per_second": 2.679
|
| 1409 |
+
}
|
| 1410 |
+
],
|
| 1411 |
+
"logging_steps": 1000,
|
| 1412 |
+
"max_steps": 120000,
|
| 1413 |
+
"num_input_tokens_seen": 0,
|
| 1414 |
+
"num_train_epochs": 17,
|
| 1415 |
+
"save_steps": 10000,
|
| 1416 |
+
"stateful_callbacks": {
|
| 1417 |
+
"TrainerControl": {
|
| 1418 |
+
"args": {
|
| 1419 |
+
"should_epoch_stop": false,
|
| 1420 |
+
"should_evaluate": false,
|
| 1421 |
+
"should_log": false,
|
| 1422 |
+
"should_save": true,
|
| 1423 |
+
"should_training_stop": true
|
| 1424 |
+
},
|
| 1425 |
+
"attributes": {}
|
| 1426 |
+
}
|
| 1427 |
+
},
|
| 1428 |
+
"total_flos": 1.617130434771026e+19,
|
| 1429 |
+
"train_batch_size": 64,
|
| 1430 |
+
"trial_name": null,
|
| 1431 |
+
"trial_params": null
|
| 1432 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a649deb01156bb8da3ab1696f1e8bd669abf9eb2061c66e5c7389bb738bbb9f0
|
| 3 |
+
size 7377
|