ZhangYunchenY
commited on
Commit
·
d01a2d2
1
Parent(s):
588574b
[Model] roberta-base-qqp
Browse files- config.json +37 -0
- merges.txt +0 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +1 -0
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- trainer_state.json +1155 -0
- training_args.bin +3 -0
- vocab.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/mnt/lustre/zhangyunchen/transformers/roberta-base",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"RobertaForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"finetuning_task": "qqp",
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout_prob": 0.1,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"id2label": {
|
| 15 |
+
"0": "not_duplicate",
|
| 16 |
+
"1": "duplicate"
|
| 17 |
+
},
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"intermediate_size": 3072,
|
| 20 |
+
"label2id": {
|
| 21 |
+
"duplicate": 1,
|
| 22 |
+
"not_duplicate": 0
|
| 23 |
+
},
|
| 24 |
+
"layer_norm_eps": 1e-05,
|
| 25 |
+
"max_position_embeddings": 514,
|
| 26 |
+
"model_type": "roberta",
|
| 27 |
+
"num_attention_heads": 12,
|
| 28 |
+
"num_hidden_layers": 12,
|
| 29 |
+
"pad_token_id": 1,
|
| 30 |
+
"position_embedding_type": "absolute",
|
| 31 |
+
"problem_type": "single_label_classification",
|
| 32 |
+
"torch_dtype": "float32",
|
| 33 |
+
"transformers_version": "4.16.2",
|
| 34 |
+
"type_vocab_size": 1,
|
| 35 |
+
"use_cache": true,
|
| 36 |
+
"vocab_size": 50265
|
| 37 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aec343a5d7621f55ed3aab706b2cb04d8702ec27f96f1f559a45d7bed50a567e
|
| 3 |
+
size 997304739
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8199a25b34e082d1413649b2724041e1bf370d3c1983c7f9506c88c20d3f3911
|
| 3 |
+
size 498679699
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dee218b761bdf36b7ff8ae49e813211784edea372cb7a4709df6e6cef1ba8c4e
|
| 3 |
+
size 14659
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6aa4f49d70d3829f1bc7776e372c9d7f2457a7f12aef9e49f2d954bae5f477b8
|
| 3 |
+
size 623
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": false}}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "add_prefix_space": false, "errors": "replace", "sep_token": "</s>", "cls_token": "<s>", "pad_token": "<pad>", "mask_token": "<mask>", "trim_offsets": true, "special_tokens_map_file": null, "name_or_path": "/mnt/lustre/zhangyunchen/transformers/roberta-base", "tokenizer_class": "RobertaTokenizer"}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1155 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9160277022013357,
|
| 3 |
+
"best_model_checkpoint": "./fp32_2e_5/models/qqp-roberta-base/checkpoint-67000",
|
| 4 |
+
"epoch": 5.892181866150734,
|
| 5 |
+
"global_step": 67000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.09,
|
| 12 |
+
"learning_rate": 2.9312619082515024e-06,
|
| 13 |
+
"loss": 0.5476,
|
| 14 |
+
"step": 1000
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.09,
|
| 18 |
+
"eval_accuracy": 0.8186000494682166,
|
| 19 |
+
"eval_combined_score": 0.781458719476822,
|
| 20 |
+
"eval_f1": 0.7443173894854274,
|
| 21 |
+
"eval_loss": 0.3999478816986084,
|
| 22 |
+
"eval_runtime": 100.5489,
|
| 23 |
+
"eval_samples_per_second": 402.093,
|
| 24 |
+
"eval_steps_per_second": 12.571,
|
| 25 |
+
"step": 1000
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.18,
|
| 29 |
+
"learning_rate": 5.862523816503005e-06,
|
| 30 |
+
"loss": 0.3783,
|
| 31 |
+
"step": 2000
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.18,
|
| 35 |
+
"eval_accuracy": 0.8439772446203314,
|
| 36 |
+
"eval_combined_score": 0.8249961873510662,
|
| 37 |
+
"eval_f1": 0.8060151300818009,
|
| 38 |
+
"eval_loss": 0.3368555009365082,
|
| 39 |
+
"eval_runtime": 105.3628,
|
| 40 |
+
"eval_samples_per_second": 383.722,
|
| 41 |
+
"eval_steps_per_second": 11.997,
|
| 42 |
+
"step": 2000
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.26,
|
| 46 |
+
"learning_rate": 8.793785724754508e-06,
|
| 47 |
+
"loss": 0.3465,
|
| 48 |
+
"step": 3000
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.26,
|
| 52 |
+
"eval_accuracy": 0.8580262181548355,
|
| 53 |
+
"eval_combined_score": 0.8396217180601624,
|
| 54 |
+
"eval_f1": 0.8212172179654894,
|
| 55 |
+
"eval_loss": 0.30989906191825867,
|
| 56 |
+
"eval_runtime": 107.6992,
|
| 57 |
+
"eval_samples_per_second": 375.397,
|
| 58 |
+
"eval_steps_per_second": 11.736,
|
| 59 |
+
"step": 3000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.35,
|
| 63 |
+
"learning_rate": 1.172504763300601e-05,
|
| 64 |
+
"loss": 0.3284,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.35,
|
| 69 |
+
"eval_accuracy": 0.8592134553549344,
|
| 70 |
+
"eval_combined_score": 0.8436663967196221,
|
| 71 |
+
"eval_f1": 0.8281193380843096,
|
| 72 |
+
"eval_loss": 0.30866414308547974,
|
| 73 |
+
"eval_runtime": 104.8651,
|
| 74 |
+
"eval_samples_per_second": 385.543,
|
| 75 |
+
"eval_steps_per_second": 12.054,
|
| 76 |
+
"step": 4000
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.44,
|
| 80 |
+
"learning_rate": 1.4656309541257511e-05,
|
| 81 |
+
"loss": 0.3244,
|
| 82 |
+
"step": 5000
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 0.44,
|
| 86 |
+
"eval_accuracy": 0.8497897600791492,
|
| 87 |
+
"eval_combined_score": 0.835341100584005,
|
| 88 |
+
"eval_f1": 0.8208924410888607,
|
| 89 |
+
"eval_loss": 0.32785937190055847,
|
| 90 |
+
"eval_runtime": 101.6619,
|
| 91 |
+
"eval_samples_per_second": 397.691,
|
| 92 |
+
"eval_steps_per_second": 12.433,
|
| 93 |
+
"step": 5000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.53,
|
| 97 |
+
"learning_rate": 1.7587571449509016e-05,
|
| 98 |
+
"loss": 0.3087,
|
| 99 |
+
"step": 6000
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.53,
|
| 103 |
+
"eval_accuracy": 0.8763047242146921,
|
| 104 |
+
"eval_combined_score": 0.8556630011071151,
|
| 105 |
+
"eval_f1": 0.8350212779995382,
|
| 106 |
+
"eval_loss": 0.2904255986213684,
|
| 107 |
+
"eval_runtime": 100.5764,
|
| 108 |
+
"eval_samples_per_second": 401.983,
|
| 109 |
+
"eval_steps_per_second": 12.568,
|
| 110 |
+
"step": 6000
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.62,
|
| 114 |
+
"learning_rate": 1.9966880911616945e-05,
|
| 115 |
+
"loss": 0.2958,
|
| 116 |
+
"step": 7000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.62,
|
| 120 |
+
"eval_accuracy": 0.8771209497897601,
|
| 121 |
+
"eval_combined_score": 0.8548056701204515,
|
| 122 |
+
"eval_f1": 0.8324903904511429,
|
| 123 |
+
"eval_loss": 0.2824174165725708,
|
| 124 |
+
"eval_runtime": 109.1783,
|
| 125 |
+
"eval_samples_per_second": 370.312,
|
| 126 |
+
"eval_steps_per_second": 11.577,
|
| 127 |
+
"step": 7000
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 0.7,
|
| 131 |
+
"learning_rate": 1.9779767417927343e-05,
|
| 132 |
+
"loss": 0.2985,
|
| 133 |
+
"step": 8000
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 0.7,
|
| 137 |
+
"eval_accuracy": 0.8826861241652239,
|
| 138 |
+
"eval_combined_score": 0.8620260280680964,
|
| 139 |
+
"eval_f1": 0.8413659319709689,
|
| 140 |
+
"eval_loss": 0.2692563235759735,
|
| 141 |
+
"eval_runtime": 109.9785,
|
| 142 |
+
"eval_samples_per_second": 367.617,
|
| 143 |
+
"eval_steps_per_second": 11.493,
|
| 144 |
+
"step": 8000
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.79,
|
| 148 |
+
"learning_rate": 1.959265392423775e-05,
|
| 149 |
+
"loss": 0.2854,
|
| 150 |
+
"step": 9000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.79,
|
| 154 |
+
"eval_accuracy": 0.8790749443482563,
|
| 155 |
+
"eval_combined_score": 0.8634582963003614,
|
| 156 |
+
"eval_f1": 0.8478416482524664,
|
| 157 |
+
"eval_loss": 0.2772486209869385,
|
| 158 |
+
"eval_runtime": 104.6366,
|
| 159 |
+
"eval_samples_per_second": 386.385,
|
| 160 |
+
"eval_steps_per_second": 12.08,
|
| 161 |
+
"step": 9000
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.88,
|
| 165 |
+
"learning_rate": 1.940554043054815e-05,
|
| 166 |
+
"loss": 0.279,
|
| 167 |
+
"step": 10000
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.88,
|
| 171 |
+
"eval_accuracy": 0.8851842691070987,
|
| 172 |
+
"eval_combined_score": 0.8687501900384056,
|
| 173 |
+
"eval_f1": 0.8523161109697124,
|
| 174 |
+
"eval_loss": 0.26354455947875977,
|
| 175 |
+
"eval_runtime": 102.343,
|
| 176 |
+
"eval_samples_per_second": 395.044,
|
| 177 |
+
"eval_steps_per_second": 12.351,
|
| 178 |
+
"step": 10000
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.97,
|
| 182 |
+
"learning_rate": 1.9218426936858555e-05,
|
| 183 |
+
"loss": 0.2728,
|
| 184 |
+
"step": 11000
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.97,
|
| 188 |
+
"eval_accuracy": 0.8893643334157804,
|
| 189 |
+
"eval_combined_score": 0.8714862904192304,
|
| 190 |
+
"eval_f1": 0.8536082474226804,
|
| 191 |
+
"eval_loss": 0.2556052803993225,
|
| 192 |
+
"eval_runtime": 100.7949,
|
| 193 |
+
"eval_samples_per_second": 401.112,
|
| 194 |
+
"eval_steps_per_second": 12.54,
|
| 195 |
+
"step": 11000
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 1.06,
|
| 199 |
+
"learning_rate": 1.9031313443168953e-05,
|
| 200 |
+
"loss": 0.2474,
|
| 201 |
+
"step": 12000
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 1.06,
|
| 205 |
+
"eval_accuracy": 0.8884986396240415,
|
| 206 |
+
"eval_combined_score": 0.8716473743117953,
|
| 207 |
+
"eval_f1": 0.854796108999549,
|
| 208 |
+
"eval_loss": 0.26520249247550964,
|
| 209 |
+
"eval_runtime": 102.8555,
|
| 210 |
+
"eval_samples_per_second": 393.076,
|
| 211 |
+
"eval_steps_per_second": 12.289,
|
| 212 |
+
"step": 12000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 1.14,
|
| 216 |
+
"learning_rate": 1.884419994947936e-05,
|
| 217 |
+
"loss": 0.2382,
|
| 218 |
+
"step": 13000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 1.14,
|
| 222 |
+
"eval_accuracy": 0.8925302992827109,
|
| 223 |
+
"eval_combined_score": 0.8776649412356591,
|
| 224 |
+
"eval_f1": 0.8627995831886073,
|
| 225 |
+
"eval_loss": 0.25566014647483826,
|
| 226 |
+
"eval_runtime": 99.8698,
|
| 227 |
+
"eval_samples_per_second": 404.827,
|
| 228 |
+
"eval_steps_per_second": 12.656,
|
| 229 |
+
"step": 13000
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 1.23,
|
| 233 |
+
"learning_rate": 1.865708645578976e-05,
|
| 234 |
+
"loss": 0.2437,
|
| 235 |
+
"step": 14000
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 1.23,
|
| 239 |
+
"eval_accuracy": 0.8940143457828346,
|
| 240 |
+
"eval_combined_score": 0.8742345089194055,
|
| 241 |
+
"eval_f1": 0.8544546720559764,
|
| 242 |
+
"eval_loss": 0.25206559896469116,
|
| 243 |
+
"eval_runtime": 98.7673,
|
| 244 |
+
"eval_samples_per_second": 409.346,
|
| 245 |
+
"eval_steps_per_second": 12.798,
|
| 246 |
+
"step": 14000
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 1.32,
|
| 250 |
+
"learning_rate": 1.8469972962100165e-05,
|
| 251 |
+
"loss": 0.2314,
|
| 252 |
+
"step": 15000
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 1.32,
|
| 256 |
+
"eval_accuracy": 0.8962651496413554,
|
| 257 |
+
"eval_combined_score": 0.8779659416144943,
|
| 258 |
+
"eval_f1": 0.859666733587633,
|
| 259 |
+
"eval_loss": 0.2604176104068756,
|
| 260 |
+
"eval_runtime": 99.873,
|
| 261 |
+
"eval_samples_per_second": 404.814,
|
| 262 |
+
"eval_steps_per_second": 12.656,
|
| 263 |
+
"step": 15000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 1.41,
|
| 267 |
+
"learning_rate": 1.8282859468410567e-05,
|
| 268 |
+
"loss": 0.2347,
|
| 269 |
+
"step": 16000
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.41,
|
| 273 |
+
"eval_accuracy": 0.897872866683156,
|
| 274 |
+
"eval_combined_score": 0.8811078037131652,
|
| 275 |
+
"eval_f1": 0.8643427407431743,
|
| 276 |
+
"eval_loss": 0.2466404139995575,
|
| 277 |
+
"eval_runtime": 99.5334,
|
| 278 |
+
"eval_samples_per_second": 406.195,
|
| 279 |
+
"eval_steps_per_second": 12.699,
|
| 280 |
+
"step": 16000
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 1.5,
|
| 284 |
+
"learning_rate": 1.809574597472097e-05,
|
| 285 |
+
"loss": 0.2322,
|
| 286 |
+
"step": 17000
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 1.5,
|
| 290 |
+
"eval_accuracy": 0.8983675488498639,
|
| 291 |
+
"eval_combined_score": 0.8828389557469845,
|
| 292 |
+
"eval_f1": 0.867310362644105,
|
| 293 |
+
"eval_loss": 0.24636423587799072,
|
| 294 |
+
"eval_runtime": 100.203,
|
| 295 |
+
"eval_samples_per_second": 403.481,
|
| 296 |
+
"eval_steps_per_second": 12.614,
|
| 297 |
+
"step": 17000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.58,
|
| 301 |
+
"learning_rate": 1.790863248103137e-05,
|
| 302 |
+
"loss": 0.2373,
|
| 303 |
+
"step": 18000
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 1.58,
|
| 307 |
+
"eval_accuracy": 0.8971061093247589,
|
| 308 |
+
"eval_combined_score": 0.883107204810262,
|
| 309 |
+
"eval_f1": 0.8691083002957649,
|
| 310 |
+
"eval_loss": 0.2558729350566864,
|
| 311 |
+
"eval_runtime": 102.812,
|
| 312 |
+
"eval_samples_per_second": 393.242,
|
| 313 |
+
"eval_steps_per_second": 12.294,
|
| 314 |
+
"step": 18000
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 1.67,
|
| 318 |
+
"learning_rate": 1.7721518987341772e-05,
|
| 319 |
+
"loss": 0.2295,
|
| 320 |
+
"step": 19000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 1.67,
|
| 324 |
+
"eval_accuracy": 0.8999752658916647,
|
| 325 |
+
"eval_combined_score": 0.8847323563693139,
|
| 326 |
+
"eval_f1": 0.8694894468469633,
|
| 327 |
+
"eval_loss": 0.25413957238197327,
|
| 328 |
+
"eval_runtime": 101.1101,
|
| 329 |
+
"eval_samples_per_second": 399.861,
|
| 330 |
+
"eval_steps_per_second": 12.501,
|
| 331 |
+
"step": 19000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.76,
|
| 335 |
+
"learning_rate": 1.7534405493652177e-05,
|
| 336 |
+
"loss": 0.2295,
|
| 337 |
+
"step": 20000
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 1.76,
|
| 341 |
+
"eval_accuracy": 0.9001978728666832,
|
| 342 |
+
"eval_combined_score": 0.8849152590077334,
|
| 343 |
+
"eval_f1": 0.8696326451487837,
|
| 344 |
+
"eval_loss": 0.24725091457366943,
|
| 345 |
+
"eval_runtime": 102.8761,
|
| 346 |
+
"eval_samples_per_second": 392.997,
|
| 347 |
+
"eval_steps_per_second": 12.287,
|
| 348 |
+
"step": 20000
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 1.85,
|
| 352 |
+
"learning_rate": 1.734729199996258e-05,
|
| 353 |
+
"loss": 0.2281,
|
| 354 |
+
"step": 21000
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.85,
|
| 358 |
+
"eval_accuracy": 0.8977491961414791,
|
| 359 |
+
"eval_combined_score": 0.8776085657993287,
|
| 360 |
+
"eval_f1": 0.8574679354571784,
|
| 361 |
+
"eval_loss": 0.2688974142074585,
|
| 362 |
+
"eval_runtime": 99.2566,
|
| 363 |
+
"eval_samples_per_second": 407.328,
|
| 364 |
+
"eval_steps_per_second": 12.735,
|
| 365 |
+
"step": 21000
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 1.93,
|
| 369 |
+
"learning_rate": 1.716017850627298e-05,
|
| 370 |
+
"loss": 0.228,
|
| 371 |
+
"step": 22000
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 1.93,
|
| 375 |
+
"eval_accuracy": 0.9038090526836507,
|
| 376 |
+
"eval_combined_score": 0.888468647308778,
|
| 377 |
+
"eval_f1": 0.8731282419339053,
|
| 378 |
+
"eval_loss": 0.23130430281162262,
|
| 379 |
+
"eval_runtime": 99.2642,
|
| 380 |
+
"eval_samples_per_second": 407.297,
|
| 381 |
+
"eval_steps_per_second": 12.734,
|
| 382 |
+
"step": 22000
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 2.02,
|
| 386 |
+
"learning_rate": 1.6973065012583382e-05,
|
| 387 |
+
"loss": 0.2116,
|
| 388 |
+
"step": 23000
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 2.02,
|
| 392 |
+
"eval_accuracy": 0.903067029433589,
|
| 393 |
+
"eval_combined_score": 0.8880802586693867,
|
| 394 |
+
"eval_f1": 0.8730934879051844,
|
| 395 |
+
"eval_loss": 0.24339179694652557,
|
| 396 |
+
"eval_runtime": 99.8059,
|
| 397 |
+
"eval_samples_per_second": 405.086,
|
| 398 |
+
"eval_steps_per_second": 12.665,
|
| 399 |
+
"step": 23000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 2.11,
|
| 403 |
+
"learning_rate": 1.6785951518893787e-05,
|
| 404 |
+
"loss": 0.179,
|
| 405 |
+
"step": 24000
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 2.11,
|
| 409 |
+
"eval_accuracy": 0.9035617116002969,
|
| 410 |
+
"eval_combined_score": 0.8877347307631893,
|
| 411 |
+
"eval_f1": 0.8719077499260817,
|
| 412 |
+
"eval_loss": 0.2653352916240692,
|
| 413 |
+
"eval_runtime": 98.8508,
|
| 414 |
+
"eval_samples_per_second": 409.0,
|
| 415 |
+
"eval_steps_per_second": 12.787,
|
| 416 |
+
"step": 24000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 2.2,
|
| 420 |
+
"learning_rate": 1.659883802520419e-05,
|
| 421 |
+
"loss": 0.1752,
|
| 422 |
+
"step": 25000
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 2.2,
|
| 426 |
+
"eval_accuracy": 0.9018303240168192,
|
| 427 |
+
"eval_combined_score": 0.8860091930795462,
|
| 428 |
+
"eval_f1": 0.8701880621422732,
|
| 429 |
+
"eval_loss": 0.2552090883255005,
|
| 430 |
+
"eval_runtime": 99.1983,
|
| 431 |
+
"eval_samples_per_second": 407.568,
|
| 432 |
+
"eval_steps_per_second": 12.742,
|
| 433 |
+
"step": 25000
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 2.29,
|
| 437 |
+
"learning_rate": 1.6411724531514594e-05,
|
| 438 |
+
"loss": 0.1822,
|
| 439 |
+
"step": 26000
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 2.29,
|
| 443 |
+
"eval_accuracy": 0.9064803363838734,
|
| 444 |
+
"eval_combined_score": 0.8916663806141756,
|
| 445 |
+
"eval_f1": 0.8768524248444777,
|
| 446 |
+
"eval_loss": 0.2510685920715332,
|
| 447 |
+
"eval_runtime": 98.8214,
|
| 448 |
+
"eval_samples_per_second": 409.122,
|
| 449 |
+
"eval_steps_per_second": 12.791,
|
| 450 |
+
"step": 26000
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 2.37,
|
| 454 |
+
"learning_rate": 1.6224611037824992e-05,
|
| 455 |
+
"loss": 0.1826,
|
| 456 |
+
"step": 27000
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 2.37,
|
| 460 |
+
"eval_accuracy": 0.9051199604254266,
|
| 461 |
+
"eval_combined_score": 0.8899457206149071,
|
| 462 |
+
"eval_f1": 0.8747714808043876,
|
| 463 |
+
"eval_loss": 0.24112574756145477,
|
| 464 |
+
"eval_runtime": 100.4418,
|
| 465 |
+
"eval_samples_per_second": 402.521,
|
| 466 |
+
"eval_steps_per_second": 12.584,
|
| 467 |
+
"step": 27000
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 2.46,
|
| 471 |
+
"learning_rate": 1.6037497544135397e-05,
|
| 472 |
+
"loss": 0.1807,
|
| 473 |
+
"step": 28000
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.46,
|
| 477 |
+
"eval_accuracy": 0.9052683650754391,
|
| 478 |
+
"eval_combined_score": 0.8889197827506554,
|
| 479 |
+
"eval_f1": 0.8725712004258717,
|
| 480 |
+
"eval_loss": 0.28508177399635315,
|
| 481 |
+
"eval_runtime": 99.6721,
|
| 482 |
+
"eval_samples_per_second": 405.63,
|
| 483 |
+
"eval_steps_per_second": 12.682,
|
| 484 |
+
"step": 28000
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 2.55,
|
| 488 |
+
"learning_rate": 1.58503840504458e-05,
|
| 489 |
+
"loss": 0.1829,
|
| 490 |
+
"step": 29000
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 2.55,
|
| 494 |
+
"eval_accuracy": 0.903512243383626,
|
| 495 |
+
"eval_combined_score": 0.888178430111253,
|
| 496 |
+
"eval_f1": 0.8728446168388799,
|
| 497 |
+
"eval_loss": 0.24632664024829865,
|
| 498 |
+
"eval_runtime": 98.7974,
|
| 499 |
+
"eval_samples_per_second": 409.221,
|
| 500 |
+
"eval_steps_per_second": 12.794,
|
| 501 |
+
"step": 29000
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 2.64,
|
| 505 |
+
"learning_rate": 1.5663270556756204e-05,
|
| 506 |
+
"loss": 0.1832,
|
| 507 |
+
"step": 30000
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 2.64,
|
| 511 |
+
"eval_accuracy": 0.9063071976255256,
|
| 512 |
+
"eval_combined_score": 0.8893352822146093,
|
| 513 |
+
"eval_f1": 0.8723633668036931,
|
| 514 |
+
"eval_loss": 0.2510886788368225,
|
| 515 |
+
"eval_runtime": 99.841,
|
| 516 |
+
"eval_samples_per_second": 404.944,
|
| 517 |
+
"eval_steps_per_second": 12.66,
|
| 518 |
+
"step": 30000
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 2.73,
|
| 522 |
+
"learning_rate": 1.5476157063066602e-05,
|
| 523 |
+
"loss": 0.1752,
|
| 524 |
+
"step": 31000
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 2.73,
|
| 528 |
+
"eval_accuracy": 0.9047984170170665,
|
| 529 |
+
"eval_combined_score": 0.8909587334582507,
|
| 530 |
+
"eval_f1": 0.8771190498994349,
|
| 531 |
+
"eval_loss": 0.25542038679122925,
|
| 532 |
+
"eval_runtime": 98.5241,
|
| 533 |
+
"eval_samples_per_second": 410.356,
|
| 534 |
+
"eval_steps_per_second": 12.829,
|
| 535 |
+
"step": 31000
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 2.81,
|
| 539 |
+
"learning_rate": 1.5289043569377007e-05,
|
| 540 |
+
"loss": 0.1822,
|
| 541 |
+
"step": 32000
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 2.81,
|
| 545 |
+
"eval_accuracy": 0.9059361860004946,
|
| 546 |
+
"eval_combined_score": 0.8898770813542074,
|
| 547 |
+
"eval_f1": 0.87381797670792,
|
| 548 |
+
"eval_loss": 0.2384674996137619,
|
| 549 |
+
"eval_runtime": 104.6367,
|
| 550 |
+
"eval_samples_per_second": 386.384,
|
| 551 |
+
"eval_steps_per_second": 12.08,
|
| 552 |
+
"step": 32000
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 2.9,
|
| 556 |
+
"learning_rate": 1.510193007568741e-05,
|
| 557 |
+
"loss": 0.1834,
|
| 558 |
+
"step": 33000
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 2.9,
|
| 562 |
+
"eval_accuracy": 0.9094484293841207,
|
| 563 |
+
"eval_combined_score": 0.895065771906153,
|
| 564 |
+
"eval_f1": 0.8806831144281851,
|
| 565 |
+
"eval_loss": 0.25289300084114075,
|
| 566 |
+
"eval_runtime": 106.9571,
|
| 567 |
+
"eval_samples_per_second": 378.002,
|
| 568 |
+
"eval_steps_per_second": 11.818,
|
| 569 |
+
"step": 33000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 2.99,
|
| 573 |
+
"learning_rate": 1.4914816581997812e-05,
|
| 574 |
+
"loss": 0.1829,
|
| 575 |
+
"step": 34000
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 2.99,
|
| 579 |
+
"eval_accuracy": 0.9079643828839971,
|
| 580 |
+
"eval_combined_score": 0.8944023406718317,
|
| 581 |
+
"eval_f1": 0.8808402984596663,
|
| 582 |
+
"eval_loss": 0.25022605061531067,
|
| 583 |
+
"eval_runtime": 99.5394,
|
| 584 |
+
"eval_samples_per_second": 406.171,
|
| 585 |
+
"eval_steps_per_second": 12.698,
|
| 586 |
+
"step": 34000
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 3.08,
|
| 590 |
+
"learning_rate": 1.4727703088308214e-05,
|
| 591 |
+
"loss": 0.1477,
|
| 592 |
+
"step": 35000
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 3.08,
|
| 596 |
+
"eval_accuracy": 0.9075686371506307,
|
| 597 |
+
"eval_combined_score": 0.892660304050797,
|
| 598 |
+
"eval_f1": 0.8777519709509634,
|
| 599 |
+
"eval_loss": 0.27075424790382385,
|
| 600 |
+
"eval_runtime": 99.0291,
|
| 601 |
+
"eval_samples_per_second": 408.264,
|
| 602 |
+
"eval_steps_per_second": 12.764,
|
| 603 |
+
"step": 35000
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 3.17,
|
| 607 |
+
"learning_rate": 1.4540589594618617e-05,
|
| 608 |
+
"loss": 0.1429,
|
| 609 |
+
"step": 36000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 3.17,
|
| 613 |
+
"eval_accuracy": 0.908780608459065,
|
| 614 |
+
"eval_combined_score": 0.8936844838450184,
|
| 615 |
+
"eval_f1": 0.8785883592309719,
|
| 616 |
+
"eval_loss": 0.2728239893913269,
|
| 617 |
+
"eval_runtime": 99.8083,
|
| 618 |
+
"eval_samples_per_second": 405.076,
|
| 619 |
+
"eval_steps_per_second": 12.664,
|
| 620 |
+
"step": 36000
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 3.25,
|
| 624 |
+
"learning_rate": 1.435347610092902e-05,
|
| 625 |
+
"loss": 0.1424,
|
| 626 |
+
"step": 37000
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 3.25,
|
| 630 |
+
"eval_accuracy": 0.9095226317091268,
|
| 631 |
+
"eval_combined_score": 0.8944181947724124,
|
| 632 |
+
"eval_f1": 0.8793137578356978,
|
| 633 |
+
"eval_loss": 0.3181060254573822,
|
| 634 |
+
"eval_runtime": 99.1709,
|
| 635 |
+
"eval_samples_per_second": 407.68,
|
| 636 |
+
"eval_steps_per_second": 12.746,
|
| 637 |
+
"step": 37000
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 3.34,
|
| 641 |
+
"learning_rate": 1.416636260723942e-05,
|
| 642 |
+
"loss": 0.1421,
|
| 643 |
+
"step": 38000
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 3.34,
|
| 647 |
+
"eval_accuracy": 0.9079891169923324,
|
| 648 |
+
"eval_combined_score": 0.8935460697083859,
|
| 649 |
+
"eval_f1": 0.8791030224244393,
|
| 650 |
+
"eval_loss": 0.25459086894989014,
|
| 651 |
+
"eval_runtime": 101.0313,
|
| 652 |
+
"eval_samples_per_second": 400.173,
|
| 653 |
+
"eval_steps_per_second": 12.511,
|
| 654 |
+
"step": 38000
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 3.43,
|
| 658 |
+
"learning_rate": 1.3979249113549824e-05,
|
| 659 |
+
"loss": 0.1453,
|
| 660 |
+
"step": 39000
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 3.43,
|
| 664 |
+
"eval_accuracy": 0.9072965619589414,
|
| 665 |
+
"eval_combined_score": 0.8937684241287549,
|
| 666 |
+
"eval_f1": 0.8802402862985685,
|
| 667 |
+
"eval_loss": 0.2652507722377777,
|
| 668 |
+
"eval_runtime": 98.7666,
|
| 669 |
+
"eval_samples_per_second": 409.349,
|
| 670 |
+
"eval_steps_per_second": 12.798,
|
| 671 |
+
"step": 39000
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 3.52,
|
| 675 |
+
"learning_rate": 1.3792135619860227e-05,
|
| 676 |
+
"loss": 0.1442,
|
| 677 |
+
"step": 40000
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 3.52,
|
| 681 |
+
"eval_accuracy": 0.9106109324758842,
|
| 682 |
+
"eval_combined_score": 0.896522122060974,
|
| 683 |
+
"eval_f1": 0.8824333116460638,
|
| 684 |
+
"eval_loss": 0.27689746022224426,
|
| 685 |
+
"eval_runtime": 98.8771,
|
| 686 |
+
"eval_samples_per_second": 408.891,
|
| 687 |
+
"eval_steps_per_second": 12.784,
|
| 688 |
+
"step": 40000
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 3.61,
|
| 692 |
+
"learning_rate": 1.360502212617063e-05,
|
| 693 |
+
"loss": 0.1485,
|
| 694 |
+
"step": 41000
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 3.61,
|
| 698 |
+
"eval_accuracy": 0.911031412317586,
|
| 699 |
+
"eval_combined_score": 0.8962277549473212,
|
| 700 |
+
"eval_f1": 0.8814240975770563,
|
| 701 |
+
"eval_loss": 0.2877097427845001,
|
| 702 |
+
"eval_runtime": 99.1266,
|
| 703 |
+
"eval_samples_per_second": 407.862,
|
| 704 |
+
"eval_steps_per_second": 12.751,
|
| 705 |
+
"step": 41000
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 3.69,
|
| 709 |
+
"learning_rate": 1.341790863248103e-05,
|
| 710 |
+
"loss": 0.1453,
|
| 711 |
+
"step": 42000
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 3.69,
|
| 715 |
+
"eval_accuracy": 0.9098194410091516,
|
| 716 |
+
"eval_combined_score": 0.895775167500813,
|
| 717 |
+
"eval_f1": 0.8817308939924743,
|
| 718 |
+
"eval_loss": 0.27044087648391724,
|
| 719 |
+
"eval_runtime": 99.2706,
|
| 720 |
+
"eval_samples_per_second": 407.271,
|
| 721 |
+
"eval_steps_per_second": 12.733,
|
| 722 |
+
"step": 42000
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 3.78,
|
| 726 |
+
"learning_rate": 1.3230795138791434e-05,
|
| 727 |
+
"loss": 0.1492,
|
| 728 |
+
"step": 43000
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 3.78,
|
| 732 |
+
"eval_accuracy": 0.91348008904279,
|
| 733 |
+
"eval_combined_score": 0.898925685140197,
|
| 734 |
+
"eval_f1": 0.8843712812376041,
|
| 735 |
+
"eval_loss": 0.25266170501708984,
|
| 736 |
+
"eval_runtime": 100.9281,
|
| 737 |
+
"eval_samples_per_second": 400.582,
|
| 738 |
+
"eval_steps_per_second": 12.524,
|
| 739 |
+
"step": 43000
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 3.87,
|
| 743 |
+
"learning_rate": 1.3043681645101837e-05,
|
| 744 |
+
"loss": 0.1444,
|
| 745 |
+
"step": 44000
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 3.87,
|
| 749 |
+
"eval_accuracy": 0.911847637892654,
|
| 750 |
+
"eval_combined_score": 0.8970186225956724,
|
| 751 |
+
"eval_f1": 0.882189607298691,
|
| 752 |
+
"eval_loss": 0.2621923089027405,
|
| 753 |
+
"eval_runtime": 98.9247,
|
| 754 |
+
"eval_samples_per_second": 408.695,
|
| 755 |
+
"eval_steps_per_second": 12.777,
|
| 756 |
+
"step": 44000
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 3.96,
|
| 760 |
+
"learning_rate": 1.285656815141224e-05,
|
| 761 |
+
"loss": 0.1433,
|
| 762 |
+
"step": 45000
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 3.96,
|
| 766 |
+
"eval_accuracy": 0.9120949789760079,
|
| 767 |
+
"eval_combined_score": 0.8979680019732961,
|
| 768 |
+
"eval_f1": 0.8838410249705844,
|
| 769 |
+
"eval_loss": 0.27234378457069397,
|
| 770 |
+
"eval_runtime": 102.783,
|
| 771 |
+
"eval_samples_per_second": 393.353,
|
| 772 |
+
"eval_steps_per_second": 12.298,
|
| 773 |
+
"step": 45000
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 4.05,
|
| 777 |
+
"learning_rate": 1.2669454657722643e-05,
|
| 778 |
+
"loss": 0.1315,
|
| 779 |
+
"step": 46000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 4.05,
|
| 783 |
+
"eval_accuracy": 0.9134306208261193,
|
| 784 |
+
"eval_combined_score": 0.8995033318593784,
|
| 785 |
+
"eval_f1": 0.8855760428926376,
|
| 786 |
+
"eval_loss": 0.29698267579078674,
|
| 787 |
+
"eval_runtime": 104.3977,
|
| 788 |
+
"eval_samples_per_second": 387.269,
|
| 789 |
+
"eval_steps_per_second": 12.108,
|
| 790 |
+
"step": 46000
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 4.13,
|
| 794 |
+
"learning_rate": 1.2482341164033044e-05,
|
| 795 |
+
"loss": 0.1135,
|
| 796 |
+
"step": 47000
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 4.13,
|
| 800 |
+
"eval_accuracy": 0.9112540192926045,
|
| 801 |
+
"eval_combined_score": 0.8968766821639864,
|
| 802 |
+
"eval_f1": 0.8824993450353682,
|
| 803 |
+
"eval_loss": 0.30147695541381836,
|
| 804 |
+
"eval_runtime": 100.9744,
|
| 805 |
+
"eval_samples_per_second": 400.398,
|
| 806 |
+
"eval_steps_per_second": 12.518,
|
| 807 |
+
"step": 47000
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 4.22,
|
| 811 |
+
"learning_rate": 1.2295227670343448e-05,
|
| 812 |
+
"loss": 0.1178,
|
| 813 |
+
"step": 48000
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 4.22,
|
| 817 |
+
"eval_accuracy": 0.9088548107840713,
|
| 818 |
+
"eval_combined_score": 0.8948400699242306,
|
| 819 |
+
"eval_f1": 0.8808253290643899,
|
| 820 |
+
"eval_loss": 0.30377110838890076,
|
| 821 |
+
"eval_runtime": 101.3426,
|
| 822 |
+
"eval_samples_per_second": 398.944,
|
| 823 |
+
"eval_steps_per_second": 12.473,
|
| 824 |
+
"step": 48000
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 4.31,
|
| 828 |
+
"learning_rate": 1.2108114176653851e-05,
|
| 829 |
+
"loss": 0.1192,
|
| 830 |
+
"step": 49000
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 4.31,
|
| 834 |
+
"eval_accuracy": 0.9117487014593124,
|
| 835 |
+
"eval_combined_score": 0.8974361955933878,
|
| 836 |
+
"eval_f1": 0.8831236897274632,
|
| 837 |
+
"eval_loss": 0.3054348826408386,
|
| 838 |
+
"eval_runtime": 102.6949,
|
| 839 |
+
"eval_samples_per_second": 393.691,
|
| 840 |
+
"eval_steps_per_second": 12.308,
|
| 841 |
+
"step": 49000
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 4.4,
|
| 845 |
+
"learning_rate": 1.1921000682964253e-05,
|
| 846 |
+
"loss": 0.1113,
|
| 847 |
+
"step": 50000
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 4.4,
|
| 851 |
+
"eval_accuracy": 0.910784071234232,
|
| 852 |
+
"eval_combined_score": 0.8965060853118298,
|
| 853 |
+
"eval_f1": 0.8822280993894277,
|
| 854 |
+
"eval_loss": 0.31508392095565796,
|
| 855 |
+
"eval_runtime": 100.8416,
|
| 856 |
+
"eval_samples_per_second": 400.926,
|
| 857 |
+
"eval_steps_per_second": 12.535,
|
| 858 |
+
"step": 50000
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 4.49,
|
| 862 |
+
"learning_rate": 1.1733887189274656e-05,
|
| 863 |
+
"loss": 0.1176,
|
| 864 |
+
"step": 51000
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 4.49,
|
| 868 |
+
"eval_accuracy": 0.9117487014593124,
|
| 869 |
+
"eval_combined_score": 0.8964195750181319,
|
| 870 |
+
"eval_f1": 0.8810904485769512,
|
| 871 |
+
"eval_loss": 0.3410942554473877,
|
| 872 |
+
"eval_runtime": 101.1296,
|
| 873 |
+
"eval_samples_per_second": 399.784,
|
| 874 |
+
"eval_steps_per_second": 12.499,
|
| 875 |
+
"step": 51000
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 4.57,
|
| 879 |
+
"learning_rate": 1.1546773695585058e-05,
|
| 880 |
+
"loss": 0.1245,
|
| 881 |
+
"step": 52000
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 4.57,
|
| 885 |
+
"eval_accuracy": 0.9097947069008162,
|
| 886 |
+
"eval_combined_score": 0.8954127669229248,
|
| 887 |
+
"eval_f1": 0.8810308269450335,
|
| 888 |
+
"eval_loss": 0.3030332028865814,
|
| 889 |
+
"eval_runtime": 100.675,
|
| 890 |
+
"eval_samples_per_second": 401.589,
|
| 891 |
+
"eval_steps_per_second": 12.555,
|
| 892 |
+
"step": 52000
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 4.66,
|
| 896 |
+
"learning_rate": 1.1359660201895461e-05,
|
| 897 |
+
"loss": 0.1197,
|
| 898 |
+
"step": 53000
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 4.66,
|
| 902 |
+
"eval_accuracy": 0.9111056146425921,
|
| 903 |
+
"eval_combined_score": 0.8957604335940053,
|
| 904 |
+
"eval_f1": 0.8804152525454183,
|
| 905 |
+
"eval_loss": 0.3057613670825958,
|
| 906 |
+
"eval_runtime": 102.3073,
|
| 907 |
+
"eval_samples_per_second": 395.182,
|
| 908 |
+
"eval_steps_per_second": 12.355,
|
| 909 |
+
"step": 53000
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 4.75,
|
| 913 |
+
"learning_rate": 1.1172546708205863e-05,
|
| 914 |
+
"loss": 0.1226,
|
| 915 |
+
"step": 54000
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 4.75,
|
| 919 |
+
"eval_accuracy": 0.9141479099678457,
|
| 920 |
+
"eval_combined_score": 0.8996355965526801,
|
| 921 |
+
"eval_f1": 0.8851232831375145,
|
| 922 |
+
"eval_loss": 0.31216293573379517,
|
| 923 |
+
"eval_runtime": 101.2062,
|
| 924 |
+
"eval_samples_per_second": 399.481,
|
| 925 |
+
"eval_steps_per_second": 12.489,
|
| 926 |
+
"step": 54000
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 4.84,
|
| 930 |
+
"learning_rate": 1.0985433214516266e-05,
|
| 931 |
+
"loss": 0.1177,
|
| 932 |
+
"step": 55000
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 4.84,
|
| 936 |
+
"eval_accuracy": 0.9131338115260945,
|
| 937 |
+
"eval_combined_score": 0.8982512840839825,
|
| 938 |
+
"eval_f1": 0.8833687566418704,
|
| 939 |
+
"eval_loss": 0.32634615898132324,
|
| 940 |
+
"eval_runtime": 101.5822,
|
| 941 |
+
"eval_samples_per_second": 398.003,
|
| 942 |
+
"eval_steps_per_second": 12.443,
|
| 943 |
+
"step": 55000
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 4.92,
|
| 947 |
+
"learning_rate": 1.079831972082667e-05,
|
| 948 |
+
"loss": 0.1172,
|
| 949 |
+
"step": 56000
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 4.92,
|
| 953 |
+
"eval_accuracy": 0.912663863467722,
|
| 954 |
+
"eval_combined_score": 0.8966402794364677,
|
| 955 |
+
"eval_f1": 0.8806166954052135,
|
| 956 |
+
"eval_loss": 0.3014249801635742,
|
| 957 |
+
"eval_runtime": 101.6983,
|
| 958 |
+
"eval_samples_per_second": 397.549,
|
| 959 |
+
"eval_steps_per_second": 12.429,
|
| 960 |
+
"step": 56000
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 5.01,
|
| 964 |
+
"learning_rate": 1.0611206227137073e-05,
|
| 965 |
+
"loss": 0.1123,
|
| 966 |
+
"step": 57000
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 5.01,
|
| 970 |
+
"eval_accuracy": 0.914296314617858,
|
| 971 |
+
"eval_combined_score": 0.8996530522745814,
|
| 972 |
+
"eval_f1": 0.8850097899313046,
|
| 973 |
+
"eval_loss": 0.3819635510444641,
|
| 974 |
+
"eval_runtime": 101.5331,
|
| 975 |
+
"eval_samples_per_second": 398.195,
|
| 976 |
+
"eval_steps_per_second": 12.449,
|
| 977 |
+
"step": 57000
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 5.1,
|
| 981 |
+
"learning_rate": 1.0424092733447473e-05,
|
| 982 |
+
"loss": 0.0951,
|
| 983 |
+
"step": 58000
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 5.1,
|
| 987 |
+
"eval_accuracy": 0.9105614642592135,
|
| 988 |
+
"eval_combined_score": 0.8972352717469176,
|
| 989 |
+
"eval_f1": 0.8839090792346218,
|
| 990 |
+
"eval_loss": 0.3715880215167999,
|
| 991 |
+
"eval_runtime": 101.3316,
|
| 992 |
+
"eval_samples_per_second": 398.987,
|
| 993 |
+
"eval_steps_per_second": 12.474,
|
| 994 |
+
"step": 58000
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 5.19,
|
| 998 |
+
"learning_rate": 1.0236979239757876e-05,
|
| 999 |
+
"loss": 0.0991,
|
| 1000 |
+
"step": 59000
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 5.19,
|
| 1004 |
+
"eval_accuracy": 0.9139995053178332,
|
| 1005 |
+
"eval_combined_score": 0.8992518702473571,
|
| 1006 |
+
"eval_f1": 0.8845042351768809,
|
| 1007 |
+
"eval_loss": 0.37594878673553467,
|
| 1008 |
+
"eval_runtime": 101.3841,
|
| 1009 |
+
"eval_samples_per_second": 398.781,
|
| 1010 |
+
"eval_steps_per_second": 12.467,
|
| 1011 |
+
"step": 59000
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 5.28,
|
| 1015 |
+
"learning_rate": 1.004986574606828e-05,
|
| 1016 |
+
"loss": 0.1035,
|
| 1017 |
+
"step": 60000
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 5.28,
|
| 1021 |
+
"eval_accuracy": 0.9140242394261687,
|
| 1022 |
+
"eval_combined_score": 0.8998636256975642,
|
| 1023 |
+
"eval_f1": 0.8857030119689596,
|
| 1024 |
+
"eval_loss": 0.3245335817337036,
|
| 1025 |
+
"eval_runtime": 102.513,
|
| 1026 |
+
"eval_samples_per_second": 394.389,
|
| 1027 |
+
"eval_steps_per_second": 12.33,
|
| 1028 |
+
"step": 60000
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 5.36,
|
| 1032 |
+
"learning_rate": 9.862752252378681e-06,
|
| 1033 |
+
"loss": 0.0967,
|
| 1034 |
+
"step": 61000
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 5.36,
|
| 1038 |
+
"eval_accuracy": 0.9130101409844175,
|
| 1039 |
+
"eval_combined_score": 0.8977728972181029,
|
| 1040 |
+
"eval_f1": 0.8825356534517885,
|
| 1041 |
+
"eval_loss": 0.3583969473838806,
|
| 1042 |
+
"eval_runtime": 104.7234,
|
| 1043 |
+
"eval_samples_per_second": 386.065,
|
| 1044 |
+
"eval_steps_per_second": 12.07,
|
| 1045 |
+
"step": 61000
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 5.45,
|
| 1049 |
+
"learning_rate": 9.675638758689083e-06,
|
| 1050 |
+
"loss": 0.1001,
|
| 1051 |
+
"step": 62000
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 5.45,
|
| 1055 |
+
"eval_accuracy": 0.9139995053178332,
|
| 1056 |
+
"eval_combined_score": 0.8993361485592781,
|
| 1057 |
+
"eval_f1": 0.8846727918007231,
|
| 1058 |
+
"eval_loss": 0.3541596829891205,
|
| 1059 |
+
"eval_runtime": 101.634,
|
| 1060 |
+
"eval_samples_per_second": 397.8,
|
| 1061 |
+
"eval_steps_per_second": 12.437,
|
| 1062 |
+
"step": 62000
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 5.54,
|
| 1066 |
+
"learning_rate": 9.488525264999486e-06,
|
| 1067 |
+
"loss": 0.0999,
|
| 1068 |
+
"step": 63000
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 5.54,
|
| 1072 |
+
"eval_accuracy": 0.9132327479594361,
|
| 1073 |
+
"eval_combined_score": 0.8982896913513594,
|
| 1074 |
+
"eval_f1": 0.8833466347432828,
|
| 1075 |
+
"eval_loss": 0.36085495352745056,
|
| 1076 |
+
"eval_runtime": 103.5986,
|
| 1077 |
+
"eval_samples_per_second": 390.256,
|
| 1078 |
+
"eval_steps_per_second": 12.201,
|
| 1079 |
+
"step": 63000
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 5.63,
|
| 1083 |
+
"learning_rate": 9.301411771309888e-06,
|
| 1084 |
+
"loss": 0.0961,
|
| 1085 |
+
"step": 64000
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 5.63,
|
| 1089 |
+
"eval_accuracy": 0.9137026960178085,
|
| 1090 |
+
"eval_combined_score": 0.8989772147109578,
|
| 1091 |
+
"eval_f1": 0.884251733404107,
|
| 1092 |
+
"eval_loss": 0.3343105912208557,
|
| 1093 |
+
"eval_runtime": 101.6502,
|
| 1094 |
+
"eval_samples_per_second": 397.736,
|
| 1095 |
+
"eval_steps_per_second": 12.435,
|
| 1096 |
+
"step": 64000
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 5.72,
|
| 1100 |
+
"learning_rate": 9.114298277620291e-06,
|
| 1101 |
+
"loss": 0.0992,
|
| 1102 |
+
"step": 65000
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 5.72,
|
| 1106 |
+
"eval_accuracy": 0.9153846153846154,
|
| 1107 |
+
"eval_combined_score": 0.9020065310865193,
|
| 1108 |
+
"eval_f1": 0.8886284467884233,
|
| 1109 |
+
"eval_loss": 0.34594395756721497,
|
| 1110 |
+
"eval_runtime": 102.9906,
|
| 1111 |
+
"eval_samples_per_second": 392.56,
|
| 1112 |
+
"eval_steps_per_second": 12.273,
|
| 1113 |
+
"step": 65000
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 5.8,
|
| 1117 |
+
"learning_rate": 8.927184783930693e-06,
|
| 1118 |
+
"loss": 0.0991,
|
| 1119 |
+
"step": 66000
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 5.8,
|
| 1123 |
+
"eval_accuracy": 0.9145189215928766,
|
| 1124 |
+
"eval_combined_score": 0.9006334285510589,
|
| 1125 |
+
"eval_f1": 0.886747935509241,
|
| 1126 |
+
"eval_loss": 0.31652504205703735,
|
| 1127 |
+
"eval_runtime": 103.8354,
|
| 1128 |
+
"eval_samples_per_second": 389.366,
|
| 1129 |
+
"eval_steps_per_second": 12.173,
|
| 1130 |
+
"step": 66000
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 5.89,
|
| 1134 |
+
"learning_rate": 8.740071290241096e-06,
|
| 1135 |
+
"loss": 0.1012,
|
| 1136 |
+
"step": 67000
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 5.89,
|
| 1140 |
+
"eval_accuracy": 0.9160277022013357,
|
| 1141 |
+
"eval_combined_score": 0.9019111310689396,
|
| 1142 |
+
"eval_f1": 0.8877945599365435,
|
| 1143 |
+
"eval_loss": 0.3151375651359558,
|
| 1144 |
+
"eval_runtime": 100.9327,
|
| 1145 |
+
"eval_samples_per_second": 400.564,
|
| 1146 |
+
"eval_steps_per_second": 12.523,
|
| 1147 |
+
"step": 67000
|
| 1148 |
+
}
|
| 1149 |
+
],
|
| 1150 |
+
"max_steps": 113710,
|
| 1151 |
+
"num_train_epochs": 10,
|
| 1152 |
+
"total_flos": 1.410189745636608e+17,
|
| 1153 |
+
"trial_name": null,
|
| 1154 |
+
"trial_params": null
|
| 1155 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:623a732a7df841aca79acda727b6d6729b0dd201503704284ddd2afe22e51a0c
|
| 3 |
+
size 3055
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|