ZhangYunchenY
commited on
Commit
·
3cccf02
1
Parent(s):
d669407
[Model] roberta-base-mnli
Browse files- config.json +39 -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 +871 -0
- training_args.bin +3 -0
- vocab.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": "mnli",
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout_prob": 0.1,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"id2label": {
|
| 15 |
+
"0": "entailment",
|
| 16 |
+
"1": "neutral",
|
| 17 |
+
"2": "contradiction"
|
| 18 |
+
},
|
| 19 |
+
"initializer_range": 0.02,
|
| 20 |
+
"intermediate_size": 3072,
|
| 21 |
+
"label2id": {
|
| 22 |
+
"contradiction": 2,
|
| 23 |
+
"entailment": 0,
|
| 24 |
+
"neutral": 1
|
| 25 |
+
},
|
| 26 |
+
"layer_norm_eps": 1e-05,
|
| 27 |
+
"max_position_embeddings": 514,
|
| 28 |
+
"model_type": "roberta",
|
| 29 |
+
"num_attention_heads": 12,
|
| 30 |
+
"num_hidden_layers": 12,
|
| 31 |
+
"pad_token_id": 1,
|
| 32 |
+
"position_embedding_type": "absolute",
|
| 33 |
+
"problem_type": "single_label_classification",
|
| 34 |
+
"torch_dtype": "float32",
|
| 35 |
+
"transformers_version": "4.16.2",
|
| 36 |
+
"type_vocab_size": 1,
|
| 37 |
+
"use_cache": true,
|
| 38 |
+
"vocab_size": 50265
|
| 39 |
+
}
|
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 sha256:2333bf1907199d4cd33693f39ac301b6a9c16d4e09a5eb2f898d568936e7920a
|
| 3 |
+
size 997310499
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:66c38290eb4e5e57d2e2529380f623f372d44196391343fa63632b717fbed2e4
|
| 3 |
+
size 498682771
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:39184be35b13da29853b5ecde623e7386711bb44764acb921f2b6f92bb7c59d0
|
| 3 |
+
size 14659
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f838a3c9c26eeb4bd84b4e439ca8e2bdfde6149bcd8ae7a0b2cd01c00a8d4a31
|
| 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 +1,871 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8775343861436576,
|
| 3 |
+
"best_model_checkpoint": "./fp32/models/mnli-roberta-base/checkpoint-57000",
|
| 4 |
+
"epoch": 4.644719687092568,
|
| 5 |
+
"global_step": 57000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.08,
|
| 12 |
+
"learning_rate": 1.3579576317218903e-06,
|
| 13 |
+
"loss": 1.0971,
|
| 14 |
+
"step": 1000
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.08,
|
| 18 |
+
"eval_accuracy": 0.5194090677534386,
|
| 19 |
+
"eval_loss": 1.0465178489685059,
|
| 20 |
+
"eval_runtime": 29.1271,
|
| 21 |
+
"eval_samples_per_second": 336.972,
|
| 22 |
+
"eval_steps_per_second": 10.54,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.16,
|
| 27 |
+
"learning_rate": 2.7159152634437806e-06,
|
| 28 |
+
"loss": 0.7469,
|
| 29 |
+
"step": 2000
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.16,
|
| 33 |
+
"eval_accuracy": 0.7842078451349974,
|
| 34 |
+
"eval_loss": 0.5564178228378296,
|
| 35 |
+
"eval_runtime": 27.2282,
|
| 36 |
+
"eval_samples_per_second": 360.471,
|
| 37 |
+
"eval_steps_per_second": 11.275,
|
| 38 |
+
"step": 2000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.24,
|
| 42 |
+
"learning_rate": 4.073872895165672e-06,
|
| 43 |
+
"loss": 0.5572,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.24,
|
| 48 |
+
"eval_accuracy": 0.8114111054508405,
|
| 49 |
+
"eval_loss": 0.4900236427783966,
|
| 50 |
+
"eval_runtime": 27.0307,
|
| 51 |
+
"eval_samples_per_second": 363.106,
|
| 52 |
+
"eval_steps_per_second": 11.357,
|
| 53 |
+
"step": 3000
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.33,
|
| 57 |
+
"learning_rate": 5.431830526887561e-06,
|
| 58 |
+
"loss": 0.5075,
|
| 59 |
+
"step": 4000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.33,
|
| 63 |
+
"eval_accuracy": 0.834538970962812,
|
| 64 |
+
"eval_loss": 0.4278336763381958,
|
| 65 |
+
"eval_runtime": 25.1831,
|
| 66 |
+
"eval_samples_per_second": 389.746,
|
| 67 |
+
"eval_steps_per_second": 12.191,
|
| 68 |
+
"step": 4000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.41,
|
| 72 |
+
"learning_rate": 6.7897881586094514e-06,
|
| 73 |
+
"loss": 0.4763,
|
| 74 |
+
"step": 5000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.41,
|
| 78 |
+
"eval_accuracy": 0.8364747834946511,
|
| 79 |
+
"eval_loss": 0.41645798087120056,
|
| 80 |
+
"eval_runtime": 26.3531,
|
| 81 |
+
"eval_samples_per_second": 372.441,
|
| 82 |
+
"eval_steps_per_second": 11.649,
|
| 83 |
+
"step": 5000
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.49,
|
| 87 |
+
"learning_rate": 8.147745790331343e-06,
|
| 88 |
+
"loss": 0.4613,
|
| 89 |
+
"step": 6000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.49,
|
| 93 |
+
"eval_accuracy": 0.8478858889454916,
|
| 94 |
+
"eval_loss": 0.39705243706703186,
|
| 95 |
+
"eval_runtime": 27.5189,
|
| 96 |
+
"eval_samples_per_second": 356.663,
|
| 97 |
+
"eval_steps_per_second": 11.156,
|
| 98 |
+
"step": 6000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.57,
|
| 102 |
+
"learning_rate": 9.505703422053234e-06,
|
| 103 |
+
"loss": 0.4539,
|
| 104 |
+
"step": 7000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.57,
|
| 108 |
+
"eval_accuracy": 0.8448293428425879,
|
| 109 |
+
"eval_loss": 0.40062254667282104,
|
| 110 |
+
"eval_runtime": 25.5016,
|
| 111 |
+
"eval_samples_per_second": 384.878,
|
| 112 |
+
"eval_steps_per_second": 12.038,
|
| 113 |
+
"step": 7000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.65,
|
| 117 |
+
"learning_rate": 9.94486632684906e-06,
|
| 118 |
+
"loss": 0.4441,
|
| 119 |
+
"step": 8000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.65,
|
| 123 |
+
"eval_accuracy": 0.8479877738155884,
|
| 124 |
+
"eval_loss": 0.39510154724121094,
|
| 125 |
+
"eval_runtime": 26.4014,
|
| 126 |
+
"eval_samples_per_second": 371.76,
|
| 127 |
+
"eval_steps_per_second": 11.628,
|
| 128 |
+
"step": 8000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.73,
|
| 132 |
+
"learning_rate": 9.85817816151739e-06,
|
| 133 |
+
"loss": 0.4293,
|
| 134 |
+
"step": 9000
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.73,
|
| 138 |
+
"eval_accuracy": 0.8596026490066225,
|
| 139 |
+
"eval_loss": 0.3864976763725281,
|
| 140 |
+
"eval_runtime": 27.4299,
|
| 141 |
+
"eval_samples_per_second": 357.821,
|
| 142 |
+
"eval_steps_per_second": 11.192,
|
| 143 |
+
"step": 9000
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.81,
|
| 147 |
+
"learning_rate": 9.771489996185721e-06,
|
| 148 |
+
"loss": 0.4237,
|
| 149 |
+
"step": 10000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.81,
|
| 153 |
+
"eval_accuracy": 0.8576668364747835,
|
| 154 |
+
"eval_loss": 0.3852238059043884,
|
| 155 |
+
"eval_runtime": 26.1569,
|
| 156 |
+
"eval_samples_per_second": 375.236,
|
| 157 |
+
"eval_steps_per_second": 11.737,
|
| 158 |
+
"step": 10000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.9,
|
| 162 |
+
"learning_rate": 9.684801830854052e-06,
|
| 163 |
+
"loss": 0.4214,
|
| 164 |
+
"step": 11000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.9,
|
| 168 |
+
"eval_accuracy": 0.8607233825776872,
|
| 169 |
+
"eval_loss": 0.3725791275501251,
|
| 170 |
+
"eval_runtime": 28.6257,
|
| 171 |
+
"eval_samples_per_second": 342.874,
|
| 172 |
+
"eval_steps_per_second": 10.725,
|
| 173 |
+
"step": 11000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.98,
|
| 177 |
+
"learning_rate": 9.598113665522383e-06,
|
| 178 |
+
"loss": 0.4147,
|
| 179 |
+
"step": 12000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.98,
|
| 183 |
+
"eval_accuracy": 0.862761079979623,
|
| 184 |
+
"eval_loss": 0.3637676239013672,
|
| 185 |
+
"eval_runtime": 24.8449,
|
| 186 |
+
"eval_samples_per_second": 395.052,
|
| 187 |
+
"eval_steps_per_second": 12.357,
|
| 188 |
+
"step": 12000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 1.06,
|
| 192 |
+
"learning_rate": 9.511425500190714e-06,
|
| 193 |
+
"loss": 0.3707,
|
| 194 |
+
"step": 13000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 1.06,
|
| 198 |
+
"eval_accuracy": 0.8629648497198166,
|
| 199 |
+
"eval_loss": 0.3756280243396759,
|
| 200 |
+
"eval_runtime": 28.9144,
|
| 201 |
+
"eval_samples_per_second": 339.451,
|
| 202 |
+
"eval_steps_per_second": 10.618,
|
| 203 |
+
"step": 13000
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 1.14,
|
| 207 |
+
"learning_rate": 9.424737334859045e-06,
|
| 208 |
+
"loss": 0.3565,
|
| 209 |
+
"step": 14000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 1.14,
|
| 213 |
+
"eval_accuracy": 0.8644931227712684,
|
| 214 |
+
"eval_loss": 0.3675437867641449,
|
| 215 |
+
"eval_runtime": 28.4842,
|
| 216 |
+
"eval_samples_per_second": 344.578,
|
| 217 |
+
"eval_steps_per_second": 10.778,
|
| 218 |
+
"step": 14000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 1.22,
|
| 222 |
+
"learning_rate": 9.338049169527376e-06,
|
| 223 |
+
"loss": 0.3662,
|
| 224 |
+
"step": 15000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 1.22,
|
| 228 |
+
"eval_accuracy": 0.866225165562914,
|
| 229 |
+
"eval_loss": 0.3639264404773712,
|
| 230 |
+
"eval_runtime": 29.9404,
|
| 231 |
+
"eval_samples_per_second": 327.818,
|
| 232 |
+
"eval_steps_per_second": 10.254,
|
| 233 |
+
"step": 15000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 1.3,
|
| 237 |
+
"learning_rate": 9.251361004195707e-06,
|
| 238 |
+
"loss": 0.3561,
|
| 239 |
+
"step": 16000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 1.3,
|
| 243 |
+
"eval_accuracy": 0.8615384615384616,
|
| 244 |
+
"eval_loss": 0.36105281114578247,
|
| 245 |
+
"eval_runtime": 24.6832,
|
| 246 |
+
"eval_samples_per_second": 397.639,
|
| 247 |
+
"eval_steps_per_second": 12.438,
|
| 248 |
+
"step": 16000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.39,
|
| 252 |
+
"learning_rate": 9.16467283886404e-06,
|
| 253 |
+
"loss": 0.3605,
|
| 254 |
+
"step": 17000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 1.39,
|
| 258 |
+
"eval_accuracy": 0.8665308201732043,
|
| 259 |
+
"eval_loss": 0.3572224974632263,
|
| 260 |
+
"eval_runtime": 25.556,
|
| 261 |
+
"eval_samples_per_second": 384.059,
|
| 262 |
+
"eval_steps_per_second": 12.013,
|
| 263 |
+
"step": 17000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 1.47,
|
| 267 |
+
"learning_rate": 9.07798467353237e-06,
|
| 268 |
+
"loss": 0.3525,
|
| 269 |
+
"step": 18000
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.47,
|
| 273 |
+
"eval_accuracy": 0.8622516556291391,
|
| 274 |
+
"eval_loss": 0.3711968660354614,
|
| 275 |
+
"eval_runtime": 27.8478,
|
| 276 |
+
"eval_samples_per_second": 352.452,
|
| 277 |
+
"eval_steps_per_second": 11.024,
|
| 278 |
+
"step": 18000
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 1.55,
|
| 282 |
+
"learning_rate": 8.9912965082007e-06,
|
| 283 |
+
"loss": 0.3489,
|
| 284 |
+
"step": 19000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 1.55,
|
| 288 |
+
"eval_accuracy": 0.8704024452368824,
|
| 289 |
+
"eval_loss": 0.371066153049469,
|
| 290 |
+
"eval_runtime": 26.8458,
|
| 291 |
+
"eval_samples_per_second": 365.606,
|
| 292 |
+
"eval_steps_per_second": 11.436,
|
| 293 |
+
"step": 19000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 1.63,
|
| 297 |
+
"learning_rate": 8.904608342869031e-06,
|
| 298 |
+
"loss": 0.3553,
|
| 299 |
+
"step": 20000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 1.63,
|
| 303 |
+
"eval_accuracy": 0.8635761589403973,
|
| 304 |
+
"eval_loss": 0.365778386592865,
|
| 305 |
+
"eval_runtime": 27.8617,
|
| 306 |
+
"eval_samples_per_second": 352.276,
|
| 307 |
+
"eval_steps_per_second": 11.019,
|
| 308 |
+
"step": 20000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 1.71,
|
| 312 |
+
"learning_rate": 8.817920177537364e-06,
|
| 313 |
+
"loss": 0.3478,
|
| 314 |
+
"step": 21000
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 1.71,
|
| 318 |
+
"eval_accuracy": 0.8650025471217524,
|
| 319 |
+
"eval_loss": 0.365774542093277,
|
| 320 |
+
"eval_runtime": 25.6814,
|
| 321 |
+
"eval_samples_per_second": 382.183,
|
| 322 |
+
"eval_steps_per_second": 11.954,
|
| 323 |
+
"step": 21000
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 1.79,
|
| 327 |
+
"learning_rate": 8.731232012205695e-06,
|
| 328 |
+
"loss": 0.3416,
|
| 329 |
+
"step": 22000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 1.79,
|
| 333 |
+
"eval_accuracy": 0.86571574121243,
|
| 334 |
+
"eval_loss": 0.38820841908454895,
|
| 335 |
+
"eval_runtime": 25.5524,
|
| 336 |
+
"eval_samples_per_second": 384.112,
|
| 337 |
+
"eval_steps_per_second": 12.015,
|
| 338 |
+
"step": 22000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.87,
|
| 342 |
+
"learning_rate": 8.644543846874025e-06,
|
| 343 |
+
"loss": 0.3463,
|
| 344 |
+
"step": 23000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 1.87,
|
| 348 |
+
"eval_accuracy": 0.8714212939378503,
|
| 349 |
+
"eval_loss": 0.3499988317489624,
|
| 350 |
+
"eval_runtime": 27.3842,
|
| 351 |
+
"eval_samples_per_second": 358.418,
|
| 352 |
+
"eval_steps_per_second": 11.211,
|
| 353 |
+
"step": 23000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.96,
|
| 357 |
+
"learning_rate": 8.557855681542356e-06,
|
| 358 |
+
"loss": 0.3429,
|
| 359 |
+
"step": 24000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.96,
|
| 363 |
+
"eval_accuracy": 0.8645950076413652,
|
| 364 |
+
"eval_loss": 0.37212345004081726,
|
| 365 |
+
"eval_runtime": 27.0146,
|
| 366 |
+
"eval_samples_per_second": 363.322,
|
| 367 |
+
"eval_steps_per_second": 11.364,
|
| 368 |
+
"step": 24000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.04,
|
| 372 |
+
"learning_rate": 8.471167516210688e-06,
|
| 373 |
+
"loss": 0.3177,
|
| 374 |
+
"step": 25000
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 2.04,
|
| 378 |
+
"eval_accuracy": 0.8691798267957208,
|
| 379 |
+
"eval_loss": 0.37507861852645874,
|
| 380 |
+
"eval_runtime": 27.0938,
|
| 381 |
+
"eval_samples_per_second": 362.26,
|
| 382 |
+
"eval_steps_per_second": 11.331,
|
| 383 |
+
"step": 25000
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 2.12,
|
| 387 |
+
"learning_rate": 8.38447935087902e-06,
|
| 388 |
+
"loss": 0.2821,
|
| 389 |
+
"step": 26000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 2.12,
|
| 393 |
+
"eval_accuracy": 0.8716250636780438,
|
| 394 |
+
"eval_loss": 0.37794411182403564,
|
| 395 |
+
"eval_runtime": 27.0755,
|
| 396 |
+
"eval_samples_per_second": 362.505,
|
| 397 |
+
"eval_steps_per_second": 11.339,
|
| 398 |
+
"step": 26000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 2.2,
|
| 402 |
+
"learning_rate": 8.29779118554735e-06,
|
| 403 |
+
"loss": 0.2726,
|
| 404 |
+
"step": 27000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 2.2,
|
| 408 |
+
"eval_accuracy": 0.8691798267957208,
|
| 409 |
+
"eval_loss": 0.38138309121131897,
|
| 410 |
+
"eval_runtime": 28.6712,
|
| 411 |
+
"eval_samples_per_second": 342.329,
|
| 412 |
+
"eval_steps_per_second": 10.708,
|
| 413 |
+
"step": 27000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 2.28,
|
| 417 |
+
"learning_rate": 8.211103020215681e-06,
|
| 418 |
+
"loss": 0.2743,
|
| 419 |
+
"step": 28000
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 2.28,
|
| 423 |
+
"eval_accuracy": 0.8757004584819155,
|
| 424 |
+
"eval_loss": 0.3701411783695221,
|
| 425 |
+
"eval_runtime": 28.7005,
|
| 426 |
+
"eval_samples_per_second": 341.981,
|
| 427 |
+
"eval_steps_per_second": 10.697,
|
| 428 |
+
"step": 28000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 2.36,
|
| 432 |
+
"learning_rate": 8.124414854884012e-06,
|
| 433 |
+
"loss": 0.2814,
|
| 434 |
+
"step": 29000
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 2.36,
|
| 438 |
+
"eval_accuracy": 0.8675496688741722,
|
| 439 |
+
"eval_loss": 0.40641501545906067,
|
| 440 |
+
"eval_runtime": 27.6676,
|
| 441 |
+
"eval_samples_per_second": 354.747,
|
| 442 |
+
"eval_steps_per_second": 11.096,
|
| 443 |
+
"step": 29000
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 2.44,
|
| 447 |
+
"learning_rate": 8.037726689552343e-06,
|
| 448 |
+
"loss": 0.2718,
|
| 449 |
+
"step": 30000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 2.44,
|
| 453 |
+
"eval_accuracy": 0.8726439123790117,
|
| 454 |
+
"eval_loss": 0.39712002873420715,
|
| 455 |
+
"eval_runtime": 30.6579,
|
| 456 |
+
"eval_samples_per_second": 320.146,
|
| 457 |
+
"eval_steps_per_second": 10.014,
|
| 458 |
+
"step": 30000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 2.53,
|
| 462 |
+
"learning_rate": 7.951038524220674e-06,
|
| 463 |
+
"loss": 0.2798,
|
| 464 |
+
"step": 31000
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 2.53,
|
| 468 |
+
"eval_accuracy": 0.8732552215995925,
|
| 469 |
+
"eval_loss": 0.3579646944999695,
|
| 470 |
+
"eval_runtime": 29.1251,
|
| 471 |
+
"eval_samples_per_second": 336.994,
|
| 472 |
+
"eval_steps_per_second": 10.541,
|
| 473 |
+
"step": 31000
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.61,
|
| 477 |
+
"learning_rate": 7.864350358889005e-06,
|
| 478 |
+
"loss": 0.2808,
|
| 479 |
+
"step": 32000
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 2.61,
|
| 483 |
+
"eval_accuracy": 0.8733571064696892,
|
| 484 |
+
"eval_loss": 0.3730563819408417,
|
| 485 |
+
"eval_runtime": 28.2414,
|
| 486 |
+
"eval_samples_per_second": 347.54,
|
| 487 |
+
"eval_steps_per_second": 10.871,
|
| 488 |
+
"step": 32000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 2.69,
|
| 492 |
+
"learning_rate": 7.777662193557336e-06,
|
| 493 |
+
"loss": 0.2854,
|
| 494 |
+
"step": 33000
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 2.69,
|
| 498 |
+
"eval_accuracy": 0.8724401426388181,
|
| 499 |
+
"eval_loss": 0.3838872015476227,
|
| 500 |
+
"eval_runtime": 28.3639,
|
| 501 |
+
"eval_samples_per_second": 346.039,
|
| 502 |
+
"eval_steps_per_second": 10.824,
|
| 503 |
+
"step": 33000
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 2.77,
|
| 507 |
+
"learning_rate": 7.690974028225667e-06,
|
| 508 |
+
"loss": 0.2802,
|
| 509 |
+
"step": 34000
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 2.77,
|
| 513 |
+
"eval_accuracy": 0.8763117677024962,
|
| 514 |
+
"eval_loss": 0.3689836263656616,
|
| 515 |
+
"eval_runtime": 31.7713,
|
| 516 |
+
"eval_samples_per_second": 308.927,
|
| 517 |
+
"eval_steps_per_second": 9.663,
|
| 518 |
+
"step": 34000
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 2.85,
|
| 522 |
+
"learning_rate": 7.604285862893998e-06,
|
| 523 |
+
"loss": 0.2825,
|
| 524 |
+
"step": 35000
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 2.85,
|
| 528 |
+
"eval_accuracy": 0.872032603158431,
|
| 529 |
+
"eval_loss": 0.37332993745803833,
|
| 530 |
+
"eval_runtime": 28.2862,
|
| 531 |
+
"eval_samples_per_second": 346.99,
|
| 532 |
+
"eval_steps_per_second": 10.853,
|
| 533 |
+
"step": 35000
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 2.93,
|
| 537 |
+
"learning_rate": 7.51759769756233e-06,
|
| 538 |
+
"loss": 0.2811,
|
| 539 |
+
"step": 36000
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 2.93,
|
| 543 |
+
"eval_accuracy": 0.872032603158431,
|
| 544 |
+
"eval_loss": 0.3704814016819,
|
| 545 |
+
"eval_runtime": 27.6376,
|
| 546 |
+
"eval_samples_per_second": 355.133,
|
| 547 |
+
"eval_steps_per_second": 11.108,
|
| 548 |
+
"step": 36000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 3.01,
|
| 552 |
+
"learning_rate": 7.43090953223066e-06,
|
| 553 |
+
"loss": 0.2733,
|
| 554 |
+
"step": 37000
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 3.01,
|
| 558 |
+
"eval_accuracy": 0.8758023433520122,
|
| 559 |
+
"eval_loss": 0.4095652401447296,
|
| 560 |
+
"eval_runtime": 26.7806,
|
| 561 |
+
"eval_samples_per_second": 366.496,
|
| 562 |
+
"eval_steps_per_second": 11.464,
|
| 563 |
+
"step": 37000
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 3.1,
|
| 567 |
+
"learning_rate": 7.344221366898991e-06,
|
| 568 |
+
"loss": 0.2186,
|
| 569 |
+
"step": 38000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 3.1,
|
| 573 |
+
"eval_accuracy": 0.8766174223127865,
|
| 574 |
+
"eval_loss": 0.4097856283187866,
|
| 575 |
+
"eval_runtime": 30.263,
|
| 576 |
+
"eval_samples_per_second": 324.324,
|
| 577 |
+
"eval_steps_per_second": 10.144,
|
| 578 |
+
"step": 38000
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 3.18,
|
| 582 |
+
"learning_rate": 7.257533201567322e-06,
|
| 583 |
+
"loss": 0.2228,
|
| 584 |
+
"step": 39000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 3.18,
|
| 588 |
+
"eval_accuracy": 0.8757004584819155,
|
| 589 |
+
"eval_loss": 0.41501158475875854,
|
| 590 |
+
"eval_runtime": 27.8042,
|
| 591 |
+
"eval_samples_per_second": 353.004,
|
| 592 |
+
"eval_steps_per_second": 11.042,
|
| 593 |
+
"step": 39000
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 3.26,
|
| 597 |
+
"learning_rate": 7.170845036235654e-06,
|
| 598 |
+
"loss": 0.2247,
|
| 599 |
+
"step": 40000
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 3.26,
|
| 603 |
+
"eval_accuracy": 0.8736627610799796,
|
| 604 |
+
"eval_loss": 0.4023166596889496,
|
| 605 |
+
"eval_runtime": 28.6307,
|
| 606 |
+
"eval_samples_per_second": 342.813,
|
| 607 |
+
"eval_steps_per_second": 10.723,
|
| 608 |
+
"step": 40000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 3.34,
|
| 612 |
+
"learning_rate": 7.084156870903984e-06,
|
| 613 |
+
"loss": 0.2236,
|
| 614 |
+
"step": 41000
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 3.34,
|
| 618 |
+
"eval_accuracy": 0.873458991339786,
|
| 619 |
+
"eval_loss": 0.401944100856781,
|
| 620 |
+
"eval_runtime": 42.8415,
|
| 621 |
+
"eval_samples_per_second": 229.1,
|
| 622 |
+
"eval_steps_per_second": 7.166,
|
| 623 |
+
"step": 41000
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 3.42,
|
| 627 |
+
"learning_rate": 6.997468705572315e-06,
|
| 628 |
+
"loss": 0.2316,
|
| 629 |
+
"step": 42000
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 3.42,
|
| 633 |
+
"eval_accuracy": 0.875394803871625,
|
| 634 |
+
"eval_loss": 0.39935216307640076,
|
| 635 |
+
"eval_runtime": 27.9679,
|
| 636 |
+
"eval_samples_per_second": 350.938,
|
| 637 |
+
"eval_steps_per_second": 10.977,
|
| 638 |
+
"step": 42000
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 3.5,
|
| 642 |
+
"learning_rate": 6.910780540240647e-06,
|
| 643 |
+
"loss": 0.227,
|
| 644 |
+
"step": 43000
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 3.5,
|
| 648 |
+
"eval_accuracy": 0.8735608762098829,
|
| 649 |
+
"eval_loss": 0.40627339482307434,
|
| 650 |
+
"eval_runtime": 27.7905,
|
| 651 |
+
"eval_samples_per_second": 353.179,
|
| 652 |
+
"eval_steps_per_second": 11.047,
|
| 653 |
+
"step": 43000
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 3.59,
|
| 657 |
+
"learning_rate": 6.824092374908978e-06,
|
| 658 |
+
"loss": 0.225,
|
| 659 |
+
"step": 44000
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 3.59,
|
| 663 |
+
"eval_accuracy": 0.8738665308201732,
|
| 664 |
+
"eval_loss": 0.3787411153316498,
|
| 665 |
+
"eval_runtime": 26.634,
|
| 666 |
+
"eval_samples_per_second": 368.513,
|
| 667 |
+
"eval_steps_per_second": 11.527,
|
| 668 |
+
"step": 44000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 3.67,
|
| 672 |
+
"learning_rate": 6.737404209577309e-06,
|
| 673 |
+
"loss": 0.2213,
|
| 674 |
+
"step": 45000
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 3.67,
|
| 678 |
+
"eval_accuracy": 0.8741721854304636,
|
| 679 |
+
"eval_loss": 0.4053351879119873,
|
| 680 |
+
"eval_runtime": 26.9451,
|
| 681 |
+
"eval_samples_per_second": 364.259,
|
| 682 |
+
"eval_steps_per_second": 11.394,
|
| 683 |
+
"step": 45000
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 3.75,
|
| 687 |
+
"learning_rate": 6.6507160442456394e-06,
|
| 688 |
+
"loss": 0.228,
|
| 689 |
+
"step": 46000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 3.75,
|
| 693 |
+
"eval_accuracy": 0.8737646459500764,
|
| 694 |
+
"eval_loss": 0.4027246832847595,
|
| 695 |
+
"eval_runtime": 26.7635,
|
| 696 |
+
"eval_samples_per_second": 366.731,
|
| 697 |
+
"eval_steps_per_second": 11.471,
|
| 698 |
+
"step": 46000
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 3.83,
|
| 702 |
+
"learning_rate": 6.564027878913971e-06,
|
| 703 |
+
"loss": 0.2227,
|
| 704 |
+
"step": 47000
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 3.83,
|
| 708 |
+
"eval_accuracy": 0.87396841569027,
|
| 709 |
+
"eval_loss": 0.4178619384765625,
|
| 710 |
+
"eval_runtime": 29.4635,
|
| 711 |
+
"eval_samples_per_second": 333.124,
|
| 712 |
+
"eval_steps_per_second": 10.42,
|
| 713 |
+
"step": 47000
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 3.91,
|
| 717 |
+
"learning_rate": 6.477339713582302e-06,
|
| 718 |
+
"loss": 0.227,
|
| 719 |
+
"step": 48000
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 3.91,
|
| 723 |
+
"eval_accuracy": 0.8752929190015283,
|
| 724 |
+
"eval_loss": 0.3969513773918152,
|
| 725 |
+
"eval_runtime": 29.294,
|
| 726 |
+
"eval_samples_per_second": 335.052,
|
| 727 |
+
"eval_steps_per_second": 10.48,
|
| 728 |
+
"step": 48000
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 3.99,
|
| 732 |
+
"learning_rate": 6.390651548250633e-06,
|
| 733 |
+
"loss": 0.2257,
|
| 734 |
+
"step": 49000
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 3.99,
|
| 738 |
+
"eval_accuracy": 0.8767193071828834,
|
| 739 |
+
"eval_loss": 0.41893526911735535,
|
| 740 |
+
"eval_runtime": 30.0922,
|
| 741 |
+
"eval_samples_per_second": 326.164,
|
| 742 |
+
"eval_steps_per_second": 10.202,
|
| 743 |
+
"step": 49000
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 4.07,
|
| 747 |
+
"learning_rate": 6.303963382918964e-06,
|
| 748 |
+
"loss": 0.1808,
|
| 749 |
+
"step": 50000
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 4.07,
|
| 753 |
+
"eval_accuracy": 0.8714212939378503,
|
| 754 |
+
"eval_loss": 0.4699897766113281,
|
| 755 |
+
"eval_runtime": 27.6446,
|
| 756 |
+
"eval_samples_per_second": 355.042,
|
| 757 |
+
"eval_steps_per_second": 11.105,
|
| 758 |
+
"step": 50000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 4.16,
|
| 762 |
+
"learning_rate": 6.217275217587295e-06,
|
| 763 |
+
"loss": 0.1821,
|
| 764 |
+
"step": 51000
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 4.16,
|
| 768 |
+
"eval_accuracy": 0.87396841569027,
|
| 769 |
+
"eval_loss": 0.4464610815048218,
|
| 770 |
+
"eval_runtime": 26.6712,
|
| 771 |
+
"eval_samples_per_second": 368.001,
|
| 772 |
+
"eval_steps_per_second": 11.511,
|
| 773 |
+
"step": 51000
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 4.24,
|
| 777 |
+
"learning_rate": 6.130587052255626e-06,
|
| 778 |
+
"loss": 0.1788,
|
| 779 |
+
"step": 52000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 4.24,
|
| 783 |
+
"eval_accuracy": 0.8727457972491085,
|
| 784 |
+
"eval_loss": 0.46066877245903015,
|
| 785 |
+
"eval_runtime": 27.6226,
|
| 786 |
+
"eval_samples_per_second": 355.326,
|
| 787 |
+
"eval_steps_per_second": 11.114,
|
| 788 |
+
"step": 52000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 4.32,
|
| 792 |
+
"learning_rate": 6.043898886923958e-06,
|
| 793 |
+
"loss": 0.1822,
|
| 794 |
+
"step": 53000
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 4.32,
|
| 798 |
+
"eval_accuracy": 0.8737646459500764,
|
| 799 |
+
"eval_loss": 0.46094420552253723,
|
| 800 |
+
"eval_runtime": 27.8257,
|
| 801 |
+
"eval_samples_per_second": 352.732,
|
| 802 |
+
"eval_steps_per_second": 11.033,
|
| 803 |
+
"step": 53000
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 4.4,
|
| 807 |
+
"learning_rate": 5.957210721592289e-06,
|
| 808 |
+
"loss": 0.181,
|
| 809 |
+
"step": 54000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 4.4,
|
| 813 |
+
"eval_accuracy": 0.8774325012735609,
|
| 814 |
+
"eval_loss": 0.4538358151912689,
|
| 815 |
+
"eval_runtime": 27.7177,
|
| 816 |
+
"eval_samples_per_second": 354.106,
|
| 817 |
+
"eval_steps_per_second": 11.076,
|
| 818 |
+
"step": 54000
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 4.48,
|
| 822 |
+
"learning_rate": 5.870522556260619e-06,
|
| 823 |
+
"loss": 0.1843,
|
| 824 |
+
"step": 55000
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 4.48,
|
| 828 |
+
"eval_accuracy": 0.8711156393275599,
|
| 829 |
+
"eval_loss": 0.47502145171165466,
|
| 830 |
+
"eval_runtime": 28.0796,
|
| 831 |
+
"eval_samples_per_second": 349.543,
|
| 832 |
+
"eval_steps_per_second": 10.933,
|
| 833 |
+
"step": 55000
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 4.56,
|
| 837 |
+
"learning_rate": 5.78383439092895e-06,
|
| 838 |
+
"loss": 0.1842,
|
| 839 |
+
"step": 56000
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 4.56,
|
| 843 |
+
"eval_accuracy": 0.8742740703005604,
|
| 844 |
+
"eval_loss": 0.44675213098526,
|
| 845 |
+
"eval_runtime": 27.0482,
|
| 846 |
+
"eval_samples_per_second": 362.871,
|
| 847 |
+
"eval_steps_per_second": 11.35,
|
| 848 |
+
"step": 56000
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 4.64,
|
| 852 |
+
"learning_rate": 5.697146225597282e-06,
|
| 853 |
+
"loss": 0.1828,
|
| 854 |
+
"step": 57000
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 4.64,
|
| 858 |
+
"eval_accuracy": 0.8775343861436576,
|
| 859 |
+
"eval_loss": 0.45801690220832825,
|
| 860 |
+
"eval_runtime": 26.6517,
|
| 861 |
+
"eval_samples_per_second": 368.27,
|
| 862 |
+
"eval_steps_per_second": 11.519,
|
| 863 |
+
"step": 57000
|
| 864 |
+
}
|
| 865 |
+
],
|
| 866 |
+
"max_steps": 122720,
|
| 867 |
+
"num_train_epochs": 10,
|
| 868 |
+
"total_flos": 1.1997919225713254e+17,
|
| 869 |
+
"trial_name": null,
|
| 870 |
+
"trial_params": null
|
| 871 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:32eb09f8f1c1bc135d8b7a83ebb2056f7c2a507ccbbdb829960ed237c7e0b0d4
|
| 3 |
+
size 3055
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|