Commit
·
d7af9cb
1
Parent(s):
edfeebe
adding model finetuned on XNLI
Browse files- all_results.json +14 -0
- config.json +40 -0
- eval_results.json +9 -0
- merges.txt +0 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +1 -0
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- train_results.json +8 -0
- trainer_state.json +1231 -0
- training_args.bin +3 -0
- vocab.json +0 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_accuracy": 0.8064257028112449,
|
| 4 |
+
"eval_loss": 0.49101120233535767,
|
| 5 |
+
"eval_runtime": 1.7231,
|
| 6 |
+
"eval_samples": 2490,
|
| 7 |
+
"eval_samples_per_second": 1445.092,
|
| 8 |
+
"eval_steps_per_second": 90.536,
|
| 9 |
+
"train_loss": 0.38008860063159117,
|
| 10 |
+
"train_runtime": 3735.6095,
|
| 11 |
+
"train_samples": 392702,
|
| 12 |
+
"train_samples_per_second": 315.372,
|
| 13 |
+
"train_steps_per_second": 19.711
|
| 14 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "bertin-project/bertin-roberta-base-spanish",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"RobertaForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"finetuning_task": "xnli",
|
| 11 |
+
"gradient_checkpointing": false,
|
| 12 |
+
"hidden_act": "gelu",
|
| 13 |
+
"hidden_dropout_prob": 0.0,
|
| 14 |
+
"hidden_size": 768,
|
| 15 |
+
"id2label": {
|
| 16 |
+
"0": "LABEL_0",
|
| 17 |
+
"1": "LABEL_1",
|
| 18 |
+
"2": "LABEL_2"
|
| 19 |
+
},
|
| 20 |
+
"initializer_range": 0.02,
|
| 21 |
+
"intermediate_size": 3072,
|
| 22 |
+
"label2id": {
|
| 23 |
+
"LABEL_0": 0,
|
| 24 |
+
"LABEL_1": 1,
|
| 25 |
+
"LABEL_2": 2
|
| 26 |
+
},
|
| 27 |
+
"layer_norm_eps": 1e-05,
|
| 28 |
+
"max_position_embeddings": 514,
|
| 29 |
+
"model_type": "roberta",
|
| 30 |
+
"num_attention_heads": 12,
|
| 31 |
+
"num_hidden_layers": 12,
|
| 32 |
+
"pad_token_id": 1,
|
| 33 |
+
"position_embedding_type": "absolute",
|
| 34 |
+
"problem_type": "single_label_classification",
|
| 35 |
+
"torch_dtype": "float32",
|
| 36 |
+
"transformers_version": "4.15.0",
|
| 37 |
+
"type_vocab_size": 1,
|
| 38 |
+
"use_cache": true,
|
| 39 |
+
"vocab_size": 50262
|
| 40 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_accuracy": 0.8064257028112449,
|
| 4 |
+
"eval_loss": 0.49101120233535767,
|
| 5 |
+
"eval_runtime": 1.7231,
|
| 6 |
+
"eval_samples": 2490,
|
| 7 |
+
"eval_samples_per_second": 1445.092,
|
| 8 |
+
"eval_steps_per_second": 90.536
|
| 9 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b1d74dc02c117e2bc3819c1122ceec5516bb934d721c53e26efdcd214a5620c0
|
| 3 |
+
size 498667949
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"bos_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "eos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "unk_token": {"content": "<unk>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "sep_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "pad_token": {"content": "<pad>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "cls_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true}}
|
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": {"content": "<unk>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "bos_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "eos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "add_prefix_space": false, "errors": "replace", "sep_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "cls_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "pad_token": {"content": "<pad>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "trim_offsets": true, "max_len": 512, "special_tokens_map_file": null, "name_or_path": "bertin-project/bertin-roberta-base-spanish", "do_lower_case": false, "tokenizer_class": "RobertaTokenizer"}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"train_loss": 0.38008860063159117,
|
| 4 |
+
"train_runtime": 3735.6095,
|
| 5 |
+
"train_samples": 392702,
|
| 6 |
+
"train_samples_per_second": 315.372,
|
| 7 |
+
"train_steps_per_second": 19.711
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1231 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.49101120233535767,
|
| 3 |
+
"best_model_checkpoint": "/home/jcanete/ft-data/all_results/xnli/bertin_roberta_base/epochs_3_bs_16_lr_1e-5/checkpoint-22000",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"global_step": 73632,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.02,
|
| 12 |
+
"learning_rate": 9.932366362451109e-06,
|
| 13 |
+
"loss": 1.0444,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.04,
|
| 18 |
+
"learning_rate": 9.864461103867885e-06,
|
| 19 |
+
"loss": 0.8226,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.06,
|
| 24 |
+
"learning_rate": 9.79655584528466e-06,
|
| 25 |
+
"loss": 0.7505,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.08,
|
| 30 |
+
"learning_rate": 9.728650586701435e-06,
|
| 31 |
+
"loss": 0.7012,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.08,
|
| 36 |
+
"eval_accuracy": 0.7076305220883534,
|
| 37 |
+
"eval_loss": 0.715941309928894,
|
| 38 |
+
"eval_runtime": 1.744,
|
| 39 |
+
"eval_samples_per_second": 1427.79,
|
| 40 |
+
"eval_steps_per_second": 89.452,
|
| 41 |
+
"step": 2000
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.1,
|
| 45 |
+
"learning_rate": 9.660881138635376e-06,
|
| 46 |
+
"loss": 0.6765,
|
| 47 |
+
"step": 2500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.12,
|
| 51 |
+
"learning_rate": 9.592975880052153e-06,
|
| 52 |
+
"loss": 0.6657,
|
| 53 |
+
"step": 3000
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.14,
|
| 57 |
+
"learning_rate": 9.525070621468926e-06,
|
| 58 |
+
"loss": 0.6506,
|
| 59 |
+
"step": 3500
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.16,
|
| 63 |
+
"learning_rate": 9.457165362885704e-06,
|
| 64 |
+
"loss": 0.6259,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.16,
|
| 69 |
+
"eval_accuracy": 0.7538152610441767,
|
| 70 |
+
"eval_loss": 0.6095600724220276,
|
| 71 |
+
"eval_runtime": 1.7428,
|
| 72 |
+
"eval_samples_per_second": 1428.771,
|
| 73 |
+
"eval_steps_per_second": 89.513,
|
| 74 |
+
"step": 4000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.18,
|
| 78 |
+
"learning_rate": 9.389260104302477e-06,
|
| 79 |
+
"loss": 0.6367,
|
| 80 |
+
"step": 4500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.2,
|
| 84 |
+
"learning_rate": 9.321354845719254e-06,
|
| 85 |
+
"loss": 0.6355,
|
| 86 |
+
"step": 5000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.22,
|
| 90 |
+
"learning_rate": 9.253449587136028e-06,
|
| 91 |
+
"loss": 0.6289,
|
| 92 |
+
"step": 5500
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.24,
|
| 96 |
+
"learning_rate": 9.185544328552804e-06,
|
| 97 |
+
"loss": 0.6059,
|
| 98 |
+
"step": 6000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.24,
|
| 102 |
+
"eval_accuracy": 0.7526104417670683,
|
| 103 |
+
"eval_loss": 0.6048593521118164,
|
| 104 |
+
"eval_runtime": 1.7395,
|
| 105 |
+
"eval_samples_per_second": 1431.463,
|
| 106 |
+
"eval_steps_per_second": 89.682,
|
| 107 |
+
"step": 6000
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.26,
|
| 111 |
+
"learning_rate": 9.117639069969579e-06,
|
| 112 |
+
"loss": 0.608,
|
| 113 |
+
"step": 6500
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.29,
|
| 117 |
+
"learning_rate": 9.049869621903521e-06,
|
| 118 |
+
"loss": 0.6132,
|
| 119 |
+
"step": 7000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.31,
|
| 123 |
+
"learning_rate": 8.981964363320296e-06,
|
| 124 |
+
"loss": 0.6056,
|
| 125 |
+
"step": 7500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.33,
|
| 129 |
+
"learning_rate": 8.914059104737072e-06,
|
| 130 |
+
"loss": 0.5869,
|
| 131 |
+
"step": 8000
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.33,
|
| 135 |
+
"eval_accuracy": 0.765863453815261,
|
| 136 |
+
"eval_loss": 0.5669301152229309,
|
| 137 |
+
"eval_runtime": 1.7452,
|
| 138 |
+
"eval_samples_per_second": 1426.778,
|
| 139 |
+
"eval_steps_per_second": 89.388,
|
| 140 |
+
"step": 8000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.35,
|
| 144 |
+
"learning_rate": 8.846153846153847e-06,
|
| 145 |
+
"loss": 0.5859,
|
| 146 |
+
"step": 8500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.37,
|
| 150 |
+
"learning_rate": 8.778384398087787e-06,
|
| 151 |
+
"loss": 0.5757,
|
| 152 |
+
"step": 9000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.39,
|
| 156 |
+
"learning_rate": 8.710479139504565e-06,
|
| 157 |
+
"loss": 0.5837,
|
| 158 |
+
"step": 9500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.41,
|
| 162 |
+
"learning_rate": 8.642573880921338e-06,
|
| 163 |
+
"loss": 0.5783,
|
| 164 |
+
"step": 10000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.41,
|
| 168 |
+
"eval_accuracy": 0.7746987951807229,
|
| 169 |
+
"eval_loss": 0.5589046478271484,
|
| 170 |
+
"eval_runtime": 1.7403,
|
| 171 |
+
"eval_samples_per_second": 1430.765,
|
| 172 |
+
"eval_steps_per_second": 89.638,
|
| 173 |
+
"step": 10000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.43,
|
| 177 |
+
"learning_rate": 8.574668622338115e-06,
|
| 178 |
+
"loss": 0.5792,
|
| 179 |
+
"step": 10500
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.45,
|
| 183 |
+
"learning_rate": 8.506763363754889e-06,
|
| 184 |
+
"loss": 0.5693,
|
| 185 |
+
"step": 11000
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.47,
|
| 189 |
+
"learning_rate": 8.438858105171666e-06,
|
| 190 |
+
"loss": 0.5649,
|
| 191 |
+
"step": 11500
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.49,
|
| 195 |
+
"learning_rate": 8.371088657105606e-06,
|
| 196 |
+
"loss": 0.5557,
|
| 197 |
+
"step": 12000
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 0.49,
|
| 201 |
+
"eval_accuracy": 0.7819277108433735,
|
| 202 |
+
"eval_loss": 0.5610175132751465,
|
| 203 |
+
"eval_runtime": 1.7411,
|
| 204 |
+
"eval_samples_per_second": 1430.157,
|
| 205 |
+
"eval_steps_per_second": 89.6,
|
| 206 |
+
"step": 12000
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.51,
|
| 210 |
+
"learning_rate": 8.303183398522382e-06,
|
| 211 |
+
"loss": 0.5548,
|
| 212 |
+
"step": 12500
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.53,
|
| 216 |
+
"learning_rate": 8.235278139939157e-06,
|
| 217 |
+
"loss": 0.5766,
|
| 218 |
+
"step": 13000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.55,
|
| 222 |
+
"learning_rate": 8.167372881355933e-06,
|
| 223 |
+
"loss": 0.5551,
|
| 224 |
+
"step": 13500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.57,
|
| 228 |
+
"learning_rate": 8.099467622772708e-06,
|
| 229 |
+
"loss": 0.5614,
|
| 230 |
+
"step": 14000
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.57,
|
| 234 |
+
"eval_accuracy": 0.7819277108433735,
|
| 235 |
+
"eval_loss": 0.5594050884246826,
|
| 236 |
+
"eval_runtime": 1.7412,
|
| 237 |
+
"eval_samples_per_second": 1430.056,
|
| 238 |
+
"eval_steps_per_second": 89.594,
|
| 239 |
+
"step": 14000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.59,
|
| 243 |
+
"learning_rate": 8.031562364189484e-06,
|
| 244 |
+
"loss": 0.5486,
|
| 245 |
+
"step": 14500
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 0.61,
|
| 249 |
+
"learning_rate": 7.963657105606259e-06,
|
| 250 |
+
"loss": 0.5678,
|
| 251 |
+
"step": 15000
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 0.63,
|
| 255 |
+
"learning_rate": 7.895751847023034e-06,
|
| 256 |
+
"loss": 0.5458,
|
| 257 |
+
"step": 15500
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.65,
|
| 261 |
+
"learning_rate": 7.827982398956976e-06,
|
| 262 |
+
"loss": 0.5492,
|
| 263 |
+
"step": 16000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 0.65,
|
| 267 |
+
"eval_accuracy": 0.7927710843373494,
|
| 268 |
+
"eval_loss": 0.5257371068000793,
|
| 269 |
+
"eval_runtime": 1.7251,
|
| 270 |
+
"eval_samples_per_second": 1443.436,
|
| 271 |
+
"eval_steps_per_second": 90.432,
|
| 272 |
+
"step": 16000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.67,
|
| 276 |
+
"learning_rate": 7.76007714037375e-06,
|
| 277 |
+
"loss": 0.5466,
|
| 278 |
+
"step": 16500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.69,
|
| 282 |
+
"learning_rate": 7.692171881790527e-06,
|
| 283 |
+
"loss": 0.5364,
|
| 284 |
+
"step": 17000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.71,
|
| 288 |
+
"learning_rate": 7.624266623207302e-06,
|
| 289 |
+
"loss": 0.5526,
|
| 290 |
+
"step": 17500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.73,
|
| 294 |
+
"learning_rate": 7.556497175141243e-06,
|
| 295 |
+
"loss": 0.526,
|
| 296 |
+
"step": 18000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.73,
|
| 300 |
+
"eval_accuracy": 0.7730923694779116,
|
| 301 |
+
"eval_loss": 0.5546149611473083,
|
| 302 |
+
"eval_runtime": 1.7312,
|
| 303 |
+
"eval_samples_per_second": 1438.268,
|
| 304 |
+
"eval_steps_per_second": 90.108,
|
| 305 |
+
"step": 18000
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.75,
|
| 309 |
+
"learning_rate": 7.488591916558019e-06,
|
| 310 |
+
"loss": 0.5278,
|
| 311 |
+
"step": 18500
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.77,
|
| 315 |
+
"learning_rate": 7.420686657974794e-06,
|
| 316 |
+
"loss": 0.5469,
|
| 317 |
+
"step": 19000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.79,
|
| 321 |
+
"learning_rate": 7.35278139939157e-06,
|
| 322 |
+
"loss": 0.5406,
|
| 323 |
+
"step": 19500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 0.81,
|
| 327 |
+
"learning_rate": 7.2848761408083445e-06,
|
| 328 |
+
"loss": 0.5241,
|
| 329 |
+
"step": 20000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.81,
|
| 333 |
+
"eval_accuracy": 0.7975903614457831,
|
| 334 |
+
"eval_loss": 0.5097489953041077,
|
| 335 |
+
"eval_runtime": 1.7046,
|
| 336 |
+
"eval_samples_per_second": 1460.741,
|
| 337 |
+
"eval_steps_per_second": 91.516,
|
| 338 |
+
"step": 20000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.84,
|
| 342 |
+
"learning_rate": 7.216970882225121e-06,
|
| 343 |
+
"loss": 0.5384,
|
| 344 |
+
"step": 20500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.86,
|
| 348 |
+
"learning_rate": 7.149065623641895e-06,
|
| 349 |
+
"loss": 0.5353,
|
| 350 |
+
"step": 21000
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.88,
|
| 354 |
+
"learning_rate": 7.08116036505867e-06,
|
| 355 |
+
"loss": 0.5194,
|
| 356 |
+
"step": 21500
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.9,
|
| 360 |
+
"learning_rate": 7.013255106475446e-06,
|
| 361 |
+
"loss": 0.5358,
|
| 362 |
+
"step": 22000
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.9,
|
| 366 |
+
"eval_accuracy": 0.8064257028112449,
|
| 367 |
+
"eval_loss": 0.49101120233535767,
|
| 368 |
+
"eval_runtime": 1.7316,
|
| 369 |
+
"eval_samples_per_second": 1437.957,
|
| 370 |
+
"eval_steps_per_second": 90.089,
|
| 371 |
+
"step": 22000
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 0.92,
|
| 375 |
+
"learning_rate": 6.945485658409388e-06,
|
| 376 |
+
"loss": 0.525,
|
| 377 |
+
"step": 22500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.94,
|
| 381 |
+
"learning_rate": 6.877580399826163e-06,
|
| 382 |
+
"loss": 0.5274,
|
| 383 |
+
"step": 23000
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 0.96,
|
| 387 |
+
"learning_rate": 6.809675141242938e-06,
|
| 388 |
+
"loss": 0.5208,
|
| 389 |
+
"step": 23500
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.98,
|
| 393 |
+
"learning_rate": 6.7417698826597135e-06,
|
| 394 |
+
"loss": 0.5327,
|
| 395 |
+
"step": 24000
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.98,
|
| 399 |
+
"eval_accuracy": 0.7899598393574297,
|
| 400 |
+
"eval_loss": 0.5255107283592224,
|
| 401 |
+
"eval_runtime": 1.7081,
|
| 402 |
+
"eval_samples_per_second": 1457.767,
|
| 403 |
+
"eval_steps_per_second": 91.33,
|
| 404 |
+
"step": 24000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 1.0,
|
| 408 |
+
"learning_rate": 6.674000434593656e-06,
|
| 409 |
+
"loss": 0.5227,
|
| 410 |
+
"step": 24500
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 1.02,
|
| 414 |
+
"learning_rate": 6.606095176010431e-06,
|
| 415 |
+
"loss": 0.3802,
|
| 416 |
+
"step": 25000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.04,
|
| 420 |
+
"learning_rate": 6.538189917427205e-06,
|
| 421 |
+
"loss": 0.3898,
|
| 422 |
+
"step": 25500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.06,
|
| 426 |
+
"learning_rate": 6.470420469361148e-06,
|
| 427 |
+
"loss": 0.3598,
|
| 428 |
+
"step": 26000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 1.06,
|
| 432 |
+
"eval_accuracy": 0.797991967871486,
|
| 433 |
+
"eval_loss": 0.5192009210586548,
|
| 434 |
+
"eval_runtime": 1.7342,
|
| 435 |
+
"eval_samples_per_second": 1435.797,
|
| 436 |
+
"eval_steps_per_second": 89.954,
|
| 437 |
+
"step": 26000
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.08,
|
| 441 |
+
"learning_rate": 6.4025152107779235e-06,
|
| 442 |
+
"loss": 0.3638,
|
| 443 |
+
"step": 26500
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.1,
|
| 447 |
+
"learning_rate": 6.334609952194698e-06,
|
| 448 |
+
"loss": 0.3636,
|
| 449 |
+
"step": 27000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 1.12,
|
| 453 |
+
"learning_rate": 6.2667046936114736e-06,
|
| 454 |
+
"loss": 0.3698,
|
| 455 |
+
"step": 27500
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 1.14,
|
| 459 |
+
"learning_rate": 6.198799435028249e-06,
|
| 460 |
+
"loss": 0.3635,
|
| 461 |
+
"step": 28000
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 1.14,
|
| 465 |
+
"eval_accuracy": 0.8076305220883534,
|
| 466 |
+
"eval_loss": 0.5339061617851257,
|
| 467 |
+
"eval_runtime": 1.707,
|
| 468 |
+
"eval_samples_per_second": 1458.725,
|
| 469 |
+
"eval_steps_per_second": 91.39,
|
| 470 |
+
"step": 28000
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 1.16,
|
| 474 |
+
"learning_rate": 6.13102998696219e-06,
|
| 475 |
+
"loss": 0.3472,
|
| 476 |
+
"step": 28500
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 1.18,
|
| 480 |
+
"learning_rate": 6.063124728378966e-06,
|
| 481 |
+
"loss": 0.3599,
|
| 482 |
+
"step": 29000
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 1.2,
|
| 486 |
+
"learning_rate": 5.995219469795741e-06,
|
| 487 |
+
"loss": 0.3521,
|
| 488 |
+
"step": 29500
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 1.22,
|
| 492 |
+
"learning_rate": 5.927314211212517e-06,
|
| 493 |
+
"loss": 0.3633,
|
| 494 |
+
"step": 30000
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 1.22,
|
| 498 |
+
"eval_accuracy": 0.8040160642570281,
|
| 499 |
+
"eval_loss": 0.5154483914375305,
|
| 500 |
+
"eval_runtime": 1.7227,
|
| 501 |
+
"eval_samples_per_second": 1445.384,
|
| 502 |
+
"eval_steps_per_second": 90.554,
|
| 503 |
+
"step": 30000
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 1.24,
|
| 507 |
+
"learning_rate": 5.859408952629292e-06,
|
| 508 |
+
"loss": 0.3621,
|
| 509 |
+
"step": 30500
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 1.26,
|
| 513 |
+
"learning_rate": 5.791503694046068e-06,
|
| 514 |
+
"loss": 0.3604,
|
| 515 |
+
"step": 31000
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 1.28,
|
| 519 |
+
"learning_rate": 5.723734245980009e-06,
|
| 520 |
+
"loss": 0.355,
|
| 521 |
+
"step": 31500
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 1.3,
|
| 525 |
+
"learning_rate": 5.6558289873967845e-06,
|
| 526 |
+
"loss": 0.3586,
|
| 527 |
+
"step": 32000
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.3,
|
| 531 |
+
"eval_accuracy": 0.7975903614457831,
|
| 532 |
+
"eval_loss": 0.5583391785621643,
|
| 533 |
+
"eval_runtime": 1.7322,
|
| 534 |
+
"eval_samples_per_second": 1437.489,
|
| 535 |
+
"eval_steps_per_second": 90.06,
|
| 536 |
+
"step": 32000
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 1.32,
|
| 540 |
+
"learning_rate": 5.58792372881356e-06,
|
| 541 |
+
"loss": 0.3527,
|
| 542 |
+
"step": 32500
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.34,
|
| 546 |
+
"learning_rate": 5.520018470230335e-06,
|
| 547 |
+
"loss": 0.3644,
|
| 548 |
+
"step": 33000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.36,
|
| 552 |
+
"learning_rate": 5.452249022164276e-06,
|
| 553 |
+
"loss": 0.3769,
|
| 554 |
+
"step": 33500
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 1.39,
|
| 558 |
+
"learning_rate": 5.384343763581053e-06,
|
| 559 |
+
"loss": 0.3614,
|
| 560 |
+
"step": 34000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 1.39,
|
| 564 |
+
"eval_accuracy": 0.8056224899598393,
|
| 565 |
+
"eval_loss": 0.5339176058769226,
|
| 566 |
+
"eval_runtime": 1.706,
|
| 567 |
+
"eval_samples_per_second": 1459.585,
|
| 568 |
+
"eval_steps_per_second": 91.444,
|
| 569 |
+
"step": 34000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.41,
|
| 573 |
+
"learning_rate": 5.316438504997827e-06,
|
| 574 |
+
"loss": 0.3577,
|
| 575 |
+
"step": 34500
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 1.43,
|
| 579 |
+
"learning_rate": 5.2485332464146035e-06,
|
| 580 |
+
"loss": 0.3593,
|
| 581 |
+
"step": 35000
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 1.45,
|
| 585 |
+
"learning_rate": 5.1807637983485445e-06,
|
| 586 |
+
"loss": 0.3475,
|
| 587 |
+
"step": 35500
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 1.47,
|
| 591 |
+
"learning_rate": 5.11285853976532e-06,
|
| 592 |
+
"loss": 0.3631,
|
| 593 |
+
"step": 36000
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 1.47,
|
| 597 |
+
"eval_accuracy": 0.8076305220883534,
|
| 598 |
+
"eval_loss": 0.5045333504676819,
|
| 599 |
+
"eval_runtime": 1.7273,
|
| 600 |
+
"eval_samples_per_second": 1441.576,
|
| 601 |
+
"eval_steps_per_second": 90.316,
|
| 602 |
+
"step": 36000
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 1.49,
|
| 606 |
+
"learning_rate": 5.044953281182095e-06,
|
| 607 |
+
"loss": 0.3523,
|
| 608 |
+
"step": 36500
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 1.51,
|
| 612 |
+
"learning_rate": 4.977048022598871e-06,
|
| 613 |
+
"loss": 0.3507,
|
| 614 |
+
"step": 37000
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 1.53,
|
| 618 |
+
"learning_rate": 4.909142764015646e-06,
|
| 619 |
+
"loss": 0.3602,
|
| 620 |
+
"step": 37500
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 1.55,
|
| 624 |
+
"learning_rate": 4.841373315949587e-06,
|
| 625 |
+
"loss": 0.3542,
|
| 626 |
+
"step": 38000
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 1.55,
|
| 630 |
+
"eval_accuracy": 0.8,
|
| 631 |
+
"eval_loss": 0.5346028804779053,
|
| 632 |
+
"eval_runtime": 1.7369,
|
| 633 |
+
"eval_samples_per_second": 1433.604,
|
| 634 |
+
"eval_steps_per_second": 89.816,
|
| 635 |
+
"step": 38000
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 1.57,
|
| 639 |
+
"learning_rate": 4.773468057366363e-06,
|
| 640 |
+
"loss": 0.3527,
|
| 641 |
+
"step": 38500
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 1.59,
|
| 645 |
+
"learning_rate": 4.705562798783138e-06,
|
| 646 |
+
"loss": 0.3753,
|
| 647 |
+
"step": 39000
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.61,
|
| 651 |
+
"learning_rate": 4.6376575401999135e-06,
|
| 652 |
+
"loss": 0.3655,
|
| 653 |
+
"step": 39500
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.63,
|
| 657 |
+
"learning_rate": 4.569752281616689e-06,
|
| 658 |
+
"loss": 0.3564,
|
| 659 |
+
"step": 40000
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 1.63,
|
| 663 |
+
"eval_accuracy": 0.8008032128514057,
|
| 664 |
+
"eval_loss": 0.5383270382881165,
|
| 665 |
+
"eval_runtime": 1.7313,
|
| 666 |
+
"eval_samples_per_second": 1438.187,
|
| 667 |
+
"eval_steps_per_second": 90.103,
|
| 668 |
+
"step": 40000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.65,
|
| 672 |
+
"learning_rate": 4.501982833550631e-06,
|
| 673 |
+
"loss": 0.3574,
|
| 674 |
+
"step": 40500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.67,
|
| 678 |
+
"learning_rate": 4.434077574967406e-06,
|
| 679 |
+
"loss": 0.3566,
|
| 680 |
+
"step": 41000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 1.69,
|
| 684 |
+
"learning_rate": 4.366172316384181e-06,
|
| 685 |
+
"loss": 0.3629,
|
| 686 |
+
"step": 41500
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 1.71,
|
| 690 |
+
"learning_rate": 4.298267057800956e-06,
|
| 691 |
+
"loss": 0.3508,
|
| 692 |
+
"step": 42000
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 1.71,
|
| 696 |
+
"eval_accuracy": 0.7987951807228916,
|
| 697 |
+
"eval_loss": 0.5298879742622375,
|
| 698 |
+
"eval_runtime": 1.7287,
|
| 699 |
+
"eval_samples_per_second": 1440.391,
|
| 700 |
+
"eval_steps_per_second": 90.241,
|
| 701 |
+
"step": 42000
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 1.73,
|
| 705 |
+
"learning_rate": 4.230497609734898e-06,
|
| 706 |
+
"loss": 0.3587,
|
| 707 |
+
"step": 42500
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 1.75,
|
| 711 |
+
"learning_rate": 4.162592351151674e-06,
|
| 712 |
+
"loss": 0.3622,
|
| 713 |
+
"step": 43000
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 1.77,
|
| 717 |
+
"learning_rate": 4.094687092568449e-06,
|
| 718 |
+
"loss": 0.3599,
|
| 719 |
+
"step": 43500
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 1.79,
|
| 723 |
+
"learning_rate": 4.0267818339852245e-06,
|
| 724 |
+
"loss": 0.354,
|
| 725 |
+
"step": 44000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 1.79,
|
| 729 |
+
"eval_accuracy": 0.7927710843373494,
|
| 730 |
+
"eval_loss": 0.5385558605194092,
|
| 731 |
+
"eval_runtime": 1.733,
|
| 732 |
+
"eval_samples_per_second": 1436.852,
|
| 733 |
+
"eval_steps_per_second": 90.02,
|
| 734 |
+
"step": 44000
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 1.81,
|
| 738 |
+
"learning_rate": 3.958876575401999e-06,
|
| 739 |
+
"loss": 0.3661,
|
| 740 |
+
"step": 44500
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 1.83,
|
| 744 |
+
"learning_rate": 3.891107127335941e-06,
|
| 745 |
+
"loss": 0.3614,
|
| 746 |
+
"step": 45000
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 1.85,
|
| 750 |
+
"learning_rate": 3.823201868752716e-06,
|
| 751 |
+
"loss": 0.3561,
|
| 752 |
+
"step": 45500
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.87,
|
| 756 |
+
"learning_rate": 3.755296610169492e-06,
|
| 757 |
+
"loss": 0.3622,
|
| 758 |
+
"step": 46000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.87,
|
| 762 |
+
"eval_accuracy": 0.8100401606425702,
|
| 763 |
+
"eval_loss": 0.5162017941474915,
|
| 764 |
+
"eval_runtime": 1.7327,
|
| 765 |
+
"eval_samples_per_second": 1437.039,
|
| 766 |
+
"eval_steps_per_second": 90.031,
|
| 767 |
+
"step": 46000
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 1.89,
|
| 771 |
+
"learning_rate": 3.6873913515862668e-06,
|
| 772 |
+
"loss": 0.3472,
|
| 773 |
+
"step": 46500
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 1.91,
|
| 777 |
+
"learning_rate": 3.619621903520209e-06,
|
| 778 |
+
"loss": 0.3381,
|
| 779 |
+
"step": 47000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.94,
|
| 783 |
+
"learning_rate": 3.5517166449369845e-06,
|
| 784 |
+
"loss": 0.3443,
|
| 785 |
+
"step": 47500
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 1.96,
|
| 789 |
+
"learning_rate": 3.483811386353759e-06,
|
| 790 |
+
"loss": 0.3559,
|
| 791 |
+
"step": 48000
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 1.96,
|
| 795 |
+
"eval_accuracy": 0.8060240963855422,
|
| 796 |
+
"eval_loss": 0.513914942741394,
|
| 797 |
+
"eval_runtime": 1.7383,
|
| 798 |
+
"eval_samples_per_second": 1432.408,
|
| 799 |
+
"eval_steps_per_second": 89.741,
|
| 800 |
+
"step": 48000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.98,
|
| 804 |
+
"learning_rate": 3.4159061277705345e-06,
|
| 805 |
+
"loss": 0.3518,
|
| 806 |
+
"step": 48500
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 2.0,
|
| 810 |
+
"learning_rate": 3.34800086918731e-06,
|
| 811 |
+
"loss": 0.3554,
|
| 812 |
+
"step": 49000
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 2.02,
|
| 816 |
+
"learning_rate": 3.2803672316384183e-06,
|
| 817 |
+
"loss": 0.2133,
|
| 818 |
+
"step": 49500
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 2.04,
|
| 822 |
+
"learning_rate": 3.2124619730551937e-06,
|
| 823 |
+
"loss": 0.1896,
|
| 824 |
+
"step": 50000
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 2.04,
|
| 828 |
+
"eval_accuracy": 0.7947791164658634,
|
| 829 |
+
"eval_loss": 0.7987900972366333,
|
| 830 |
+
"eval_runtime": 1.7334,
|
| 831 |
+
"eval_samples_per_second": 1436.451,
|
| 832 |
+
"eval_steps_per_second": 89.995,
|
| 833 |
+
"step": 50000
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 2.06,
|
| 837 |
+
"learning_rate": 3.144556714471969e-06,
|
| 838 |
+
"loss": 0.1934,
|
| 839 |
+
"step": 50500
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 2.08,
|
| 843 |
+
"learning_rate": 3.0766514558887446e-06,
|
| 844 |
+
"loss": 0.1986,
|
| 845 |
+
"step": 51000
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 2.1,
|
| 849 |
+
"learning_rate": 3.008882007822686e-06,
|
| 850 |
+
"loss": 0.1973,
|
| 851 |
+
"step": 51500
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 2.12,
|
| 855 |
+
"learning_rate": 2.9409767492394614e-06,
|
| 856 |
+
"loss": 0.1867,
|
| 857 |
+
"step": 52000
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 2.12,
|
| 861 |
+
"eval_accuracy": 0.7987951807228916,
|
| 862 |
+
"eval_loss": 0.8654219508171082,
|
| 863 |
+
"eval_runtime": 1.7101,
|
| 864 |
+
"eval_samples_per_second": 1456.025,
|
| 865 |
+
"eval_steps_per_second": 91.221,
|
| 866 |
+
"step": 52000
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 2.14,
|
| 870 |
+
"learning_rate": 2.873071490656237e-06,
|
| 871 |
+
"loss": 0.2036,
|
| 872 |
+
"step": 52500
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 2.16,
|
| 876 |
+
"learning_rate": 2.8051662320730123e-06,
|
| 877 |
+
"loss": 0.1956,
|
| 878 |
+
"step": 53000
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 2.18,
|
| 882 |
+
"learning_rate": 2.737260973489787e-06,
|
| 883 |
+
"loss": 0.1934,
|
| 884 |
+
"step": 53500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 2.2,
|
| 888 |
+
"learning_rate": 2.669491525423729e-06,
|
| 889 |
+
"loss": 0.1945,
|
| 890 |
+
"step": 54000
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 2.2,
|
| 894 |
+
"eval_accuracy": 0.804417670682731,
|
| 895 |
+
"eval_loss": 0.8309808373451233,
|
| 896 |
+
"eval_runtime": 1.7106,
|
| 897 |
+
"eval_samples_per_second": 1455.613,
|
| 898 |
+
"eval_steps_per_second": 91.195,
|
| 899 |
+
"step": 54000
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 2.22,
|
| 903 |
+
"learning_rate": 2.6015862668405046e-06,
|
| 904 |
+
"loss": 0.1885,
|
| 905 |
+
"step": 54500
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 2.24,
|
| 909 |
+
"learning_rate": 2.53368100825728e-06,
|
| 910 |
+
"loss": 0.1949,
|
| 911 |
+
"step": 55000
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 2.26,
|
| 915 |
+
"learning_rate": 2.465775749674055e-06,
|
| 916 |
+
"loss": 0.2041,
|
| 917 |
+
"step": 55500
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 2.28,
|
| 921 |
+
"learning_rate": 2.39787049109083e-06,
|
| 922 |
+
"loss": 0.2005,
|
| 923 |
+
"step": 56000
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 2.28,
|
| 927 |
+
"eval_accuracy": 0.7899598393574297,
|
| 928 |
+
"eval_loss": 0.9045226573944092,
|
| 929 |
+
"eval_runtime": 1.7241,
|
| 930 |
+
"eval_samples_per_second": 1444.229,
|
| 931 |
+
"eval_steps_per_second": 90.482,
|
| 932 |
+
"step": 56000
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 2.3,
|
| 936 |
+
"learning_rate": 2.3299652325076055e-06,
|
| 937 |
+
"loss": 0.1869,
|
| 938 |
+
"step": 56500
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 2.32,
|
| 942 |
+
"learning_rate": 2.262059973924381e-06,
|
| 943 |
+
"loss": 0.1704,
|
| 944 |
+
"step": 57000
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 2.34,
|
| 948 |
+
"learning_rate": 2.1942905258583228e-06,
|
| 949 |
+
"loss": 0.197,
|
| 950 |
+
"step": 57500
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 2.36,
|
| 954 |
+
"learning_rate": 2.1263852672750978e-06,
|
| 955 |
+
"loss": 0.1879,
|
| 956 |
+
"step": 58000
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 2.36,
|
| 960 |
+
"eval_accuracy": 0.8036144578313253,
|
| 961 |
+
"eval_loss": 0.8717219233512878,
|
| 962 |
+
"eval_runtime": 1.7314,
|
| 963 |
+
"eval_samples_per_second": 1438.131,
|
| 964 |
+
"eval_steps_per_second": 90.1,
|
| 965 |
+
"step": 58000
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 2.38,
|
| 969 |
+
"learning_rate": 2.058480008691873e-06,
|
| 970 |
+
"loss": 0.1924,
|
| 971 |
+
"step": 58500
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 2.4,
|
| 975 |
+
"learning_rate": 1.9905747501086486e-06,
|
| 976 |
+
"loss": 0.2094,
|
| 977 |
+
"step": 59000
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 2.42,
|
| 981 |
+
"learning_rate": 1.9226694915254236e-06,
|
| 982 |
+
"loss": 0.1904,
|
| 983 |
+
"step": 59500
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 2.44,
|
| 987 |
+
"learning_rate": 1.8549000434593655e-06,
|
| 988 |
+
"loss": 0.2007,
|
| 989 |
+
"step": 60000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 2.44,
|
| 993 |
+
"eval_accuracy": 0.7995983935742972,
|
| 994 |
+
"eval_loss": 0.8504888415336609,
|
| 995 |
+
"eval_runtime": 1.7179,
|
| 996 |
+
"eval_samples_per_second": 1449.471,
|
| 997 |
+
"eval_steps_per_second": 90.81,
|
| 998 |
+
"step": 60000
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 2.46,
|
| 1002 |
+
"learning_rate": 1.786994784876141e-06,
|
| 1003 |
+
"loss": 0.1967,
|
| 1004 |
+
"step": 60500
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 2.49,
|
| 1008 |
+
"learning_rate": 1.7190895262929164e-06,
|
| 1009 |
+
"loss": 0.1954,
|
| 1010 |
+
"step": 61000
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 2.51,
|
| 1014 |
+
"learning_rate": 1.6511842677096914e-06,
|
| 1015 |
+
"loss": 0.1831,
|
| 1016 |
+
"step": 61500
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 2.53,
|
| 1020 |
+
"learning_rate": 1.5832790091264668e-06,
|
| 1021 |
+
"loss": 0.1861,
|
| 1022 |
+
"step": 62000
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 2.53,
|
| 1026 |
+
"eval_accuracy": 0.7983935742971887,
|
| 1027 |
+
"eval_loss": 0.8534940481185913,
|
| 1028 |
+
"eval_runtime": 1.7626,
|
| 1029 |
+
"eval_samples_per_second": 1412.7,
|
| 1030 |
+
"eval_steps_per_second": 88.507,
|
| 1031 |
+
"step": 62000
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 2.55,
|
| 1035 |
+
"learning_rate": 1.5153737505432422e-06,
|
| 1036 |
+
"loss": 0.1869,
|
| 1037 |
+
"step": 62500
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 2.57,
|
| 1041 |
+
"learning_rate": 1.4476043024771841e-06,
|
| 1042 |
+
"loss": 0.1767,
|
| 1043 |
+
"step": 63000
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 2.59,
|
| 1047 |
+
"learning_rate": 1.3796990438939591e-06,
|
| 1048 |
+
"loss": 0.1902,
|
| 1049 |
+
"step": 63500
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 2.61,
|
| 1053 |
+
"learning_rate": 1.3117937853107346e-06,
|
| 1054 |
+
"loss": 0.1836,
|
| 1055 |
+
"step": 64000
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 2.61,
|
| 1059 |
+
"eval_accuracy": 0.8,
|
| 1060 |
+
"eval_loss": 0.8850274085998535,
|
| 1061 |
+
"eval_runtime": 1.7307,
|
| 1062 |
+
"eval_samples_per_second": 1438.73,
|
| 1063 |
+
"eval_steps_per_second": 90.137,
|
| 1064 |
+
"step": 64000
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 2.63,
|
| 1068 |
+
"learning_rate": 1.2438885267275098e-06,
|
| 1069 |
+
"loss": 0.1765,
|
| 1070 |
+
"step": 64500
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 2.65,
|
| 1074 |
+
"learning_rate": 1.1761190786614516e-06,
|
| 1075 |
+
"loss": 0.1877,
|
| 1076 |
+
"step": 65000
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 2.67,
|
| 1080 |
+
"learning_rate": 1.1082138200782269e-06,
|
| 1081 |
+
"loss": 0.189,
|
| 1082 |
+
"step": 65500
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 2.69,
|
| 1086 |
+
"learning_rate": 1.0403085614950023e-06,
|
| 1087 |
+
"loss": 0.2009,
|
| 1088 |
+
"step": 66000
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 2.69,
|
| 1092 |
+
"eval_accuracy": 0.8,
|
| 1093 |
+
"eval_loss": 0.8615464568138123,
|
| 1094 |
+
"eval_runtime": 1.7274,
|
| 1095 |
+
"eval_samples_per_second": 1441.493,
|
| 1096 |
+
"eval_steps_per_second": 90.31,
|
| 1097 |
+
"step": 66000
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 2.71,
|
| 1101 |
+
"learning_rate": 9.724033029117775e-07,
|
| 1102 |
+
"loss": 0.1853,
|
| 1103 |
+
"step": 66500
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 2.73,
|
| 1107 |
+
"learning_rate": 9.044980443285528e-07,
|
| 1108 |
+
"loss": 0.1793,
|
| 1109 |
+
"step": 67000
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 2.75,
|
| 1113 |
+
"learning_rate": 8.365927857453282e-07,
|
| 1114 |
+
"loss": 0.1938,
|
| 1115 |
+
"step": 67500
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 2.77,
|
| 1119 |
+
"learning_rate": 7.686875271621035e-07,
|
| 1120 |
+
"loss": 0.1809,
|
| 1121 |
+
"step": 68000
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 2.77,
|
| 1125 |
+
"eval_accuracy": 0.7991967871485943,
|
| 1126 |
+
"eval_loss": 0.9009087085723877,
|
| 1127 |
+
"eval_runtime": 1.7248,
|
| 1128 |
+
"eval_samples_per_second": 1443.619,
|
| 1129 |
+
"eval_steps_per_second": 90.444,
|
| 1130 |
+
"step": 68000
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 2.79,
|
| 1134 |
+
"learning_rate": 7.009180790960452e-07,
|
| 1135 |
+
"loss": 0.1958,
|
| 1136 |
+
"step": 68500
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 2.81,
|
| 1140 |
+
"learning_rate": 6.330128205128206e-07,
|
| 1141 |
+
"loss": 0.1866,
|
| 1142 |
+
"step": 69000
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 2.83,
|
| 1146 |
+
"learning_rate": 5.651075619295959e-07,
|
| 1147 |
+
"loss": 0.189,
|
| 1148 |
+
"step": 69500
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 2.85,
|
| 1152 |
+
"learning_rate": 4.972023033463712e-07,
|
| 1153 |
+
"loss": 0.1885,
|
| 1154 |
+
"step": 70000
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 2.85,
|
| 1158 |
+
"eval_accuracy": 0.8064257028112449,
|
| 1159 |
+
"eval_loss": 0.89076167345047,
|
| 1160 |
+
"eval_runtime": 1.7155,
|
| 1161 |
+
"eval_samples_per_second": 1451.454,
|
| 1162 |
+
"eval_steps_per_second": 90.934,
|
| 1163 |
+
"step": 70000
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 2.87,
|
| 1167 |
+
"learning_rate": 4.292970447631465e-07,
|
| 1168 |
+
"loss": 0.1845,
|
| 1169 |
+
"step": 70500
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 2.89,
|
| 1173 |
+
"learning_rate": 3.613917861799218e-07,
|
| 1174 |
+
"loss": 0.1844,
|
| 1175 |
+
"step": 71000
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 2.91,
|
| 1179 |
+
"learning_rate": 2.936223381138636e-07,
|
| 1180 |
+
"loss": 0.1919,
|
| 1181 |
+
"step": 71500
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 2.93,
|
| 1185 |
+
"learning_rate": 2.2571707953063886e-07,
|
| 1186 |
+
"loss": 0.1941,
|
| 1187 |
+
"step": 72000
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 2.93,
|
| 1191 |
+
"eval_accuracy": 0.8036144578313253,
|
| 1192 |
+
"eval_loss": 0.8937243819236755,
|
| 1193 |
+
"eval_runtime": 1.7336,
|
| 1194 |
+
"eval_samples_per_second": 1436.277,
|
| 1195 |
+
"eval_steps_per_second": 89.984,
|
| 1196 |
+
"step": 72000
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 2.95,
|
| 1200 |
+
"learning_rate": 1.5781182094741418e-07,
|
| 1201 |
+
"loss": 0.1873,
|
| 1202 |
+
"step": 72500
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 2.97,
|
| 1206 |
+
"learning_rate": 8.990656236418949e-08,
|
| 1207 |
+
"loss": 0.1923,
|
| 1208 |
+
"step": 73000
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 2.99,
|
| 1212 |
+
"learning_rate": 2.2001303780964798e-08,
|
| 1213 |
+
"loss": 0.1831,
|
| 1214 |
+
"step": 73500
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 3.0,
|
| 1218 |
+
"step": 73632,
|
| 1219 |
+
"total_flos": 5.651008562499955e+16,
|
| 1220 |
+
"train_loss": 0.38008860063159117,
|
| 1221 |
+
"train_runtime": 3735.6095,
|
| 1222 |
+
"train_samples_per_second": 315.372,
|
| 1223 |
+
"train_steps_per_second": 19.711
|
| 1224 |
+
}
|
| 1225 |
+
],
|
| 1226 |
+
"max_steps": 73632,
|
| 1227 |
+
"num_train_epochs": 3,
|
| 1228 |
+
"total_flos": 5.651008562499955e+16,
|
| 1229 |
+
"trial_name": null,
|
| 1230 |
+
"trial_params": null
|
| 1231 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c11d664688c4e8a3091342a7be68f2c97f76b7326b065200d272d743262c7778
|
| 3 |
+
size 2991
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|