Commit
•
1b6647e
1
Parent(s):
dd493e7
adding model finetuned on XNLI
Browse files- all_results.json +14 -0
- config.json +45 -0
- eval_results.json +9 -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 +829 -0
- training_args.bin +3 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"eval_accuracy": 0.8068273092369478,
|
| 4 |
+
"eval_loss": 0.5007363557815552,
|
| 5 |
+
"eval_runtime": 1.3656,
|
| 6 |
+
"eval_samples": 2490,
|
| 7 |
+
"eval_samples_per_second": 1823.419,
|
| 8 |
+
"eval_steps_per_second": 114.238,
|
| 9 |
+
"train_loss": 0.5175694002632678,
|
| 10 |
+
"train_runtime": 2940.9235,
|
| 11 |
+
"train_samples": 392702,
|
| 12 |
+
"train_samples_per_second": 267.06,
|
| 13 |
+
"train_steps_per_second": 16.691
|
| 14 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "CenIA/albert_base_spanish",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"AlbertForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0,
|
| 7 |
+
"bos_token_id": 2,
|
| 8 |
+
"classifier_dropout_prob": 0.1,
|
| 9 |
+
"down_scale_factor": 1,
|
| 10 |
+
"embedding_size": 128,
|
| 11 |
+
"eos_token_id": 3,
|
| 12 |
+
"finetuning_task": "xnli",
|
| 13 |
+
"gap_size": 0,
|
| 14 |
+
"hidden_act": "gelu",
|
| 15 |
+
"hidden_dropout_prob": 0,
|
| 16 |
+
"hidden_size": 768,
|
| 17 |
+
"id2label": {
|
| 18 |
+
"0": "LABEL_0",
|
| 19 |
+
"1": "LABEL_1",
|
| 20 |
+
"2": "LABEL_2"
|
| 21 |
+
},
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"inner_group_num": 1,
|
| 24 |
+
"intermediate_size": 3072,
|
| 25 |
+
"label2id": {
|
| 26 |
+
"LABEL_0": 0,
|
| 27 |
+
"LABEL_1": 1,
|
| 28 |
+
"LABEL_2": 2
|
| 29 |
+
},
|
| 30 |
+
"layer_norm_eps": 1e-12,
|
| 31 |
+
"max_position_embeddings": 512,
|
| 32 |
+
"model_type": "albert",
|
| 33 |
+
"net_structure_type": 0,
|
| 34 |
+
"num_attention_heads": 12,
|
| 35 |
+
"num_hidden_groups": 1,
|
| 36 |
+
"num_hidden_layers": 12,
|
| 37 |
+
"num_memory_blocks": 0,
|
| 38 |
+
"pad_token_id": 0,
|
| 39 |
+
"position_embedding_type": "absolute",
|
| 40 |
+
"problem_type": "single_label_classification",
|
| 41 |
+
"torch_dtype": "float32",
|
| 42 |
+
"transformers_version": "4.12.5",
|
| 43 |
+
"type_vocab_size": 2,
|
| 44 |
+
"vocab_size": 31000
|
| 45 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"eval_accuracy": 0.8068273092369478,
|
| 4 |
+
"eval_loss": 0.5007363557815552,
|
| 5 |
+
"eval_runtime": 1.3656,
|
| 6 |
+
"eval_samples": 2490,
|
| 7 |
+
"eval_samples_per_second": 1823.419,
|
| 8 |
+
"eval_steps_per_second": 114.238
|
| 9 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:308ae8e38002698446682a75896060c01cdc14bdea84f718b39765d2a5607a7f
|
| 3 |
+
size 47270609
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"bos_token": "[CLS]", "eos_token": "[SEP]", "unk_token": "<unk>", "sep_token": "[SEP]", "pad_token": "<pad>", "cls_token": "[CLS]", "mask_token": "[MASK]"}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"do_lower_case": true, "remove_space": true, "keep_accents": true, "bos_token": "[CLS]", "eos_token": "[SEP]", "unk_token": "<unk>", "sep_token": "[SEP]", "pad_token": "<pad>", "cls_token": "[CLS]", "mask_token": {"content": "[MASK]", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "CenIA/albert_base_spanish", "tokenizer_class": "AlbertTokenizer"}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"train_loss": 0.5175694002632678,
|
| 4 |
+
"train_runtime": 2940.9235,
|
| 5 |
+
"train_samples": 392702,
|
| 6 |
+
"train_samples_per_second": 267.06,
|
| 7 |
+
"train_steps_per_second": 16.691
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,829 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.5007363557815552,
|
| 3 |
+
"best_model_checkpoint": "/data/jcanete/all_results/xnli/albeto_base2/epochs_2_bs_16_lr_5e-5/checkpoint-44000",
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"global_step": 49088,
|
| 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": 4.949478487614081e-05,
|
| 13 |
+
"loss": 0.8958,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.04,
|
| 18 |
+
"learning_rate": 4.8985495436766624e-05,
|
| 19 |
+
"loss": 0.7815,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.06,
|
| 24 |
+
"learning_rate": 4.847722457627119e-05,
|
| 25 |
+
"loss": 0.7453,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.08,
|
| 30 |
+
"learning_rate": 4.7967935136897005e-05,
|
| 31 |
+
"loss": 0.7032,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.08,
|
| 36 |
+
"eval_accuracy": 0.7104417670682731,
|
| 37 |
+
"eval_loss": 0.6837603449821472,
|
| 38 |
+
"eval_runtime": 3.0721,
|
| 39 |
+
"eval_samples_per_second": 810.518,
|
| 40 |
+
"eval_steps_per_second": 50.779,
|
| 41 |
+
"step": 2000
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.1,
|
| 45 |
+
"learning_rate": 4.745864569752282e-05,
|
| 46 |
+
"loss": 0.6916,
|
| 47 |
+
"step": 2500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.12,
|
| 51 |
+
"learning_rate": 4.694935625814863e-05,
|
| 52 |
+
"loss": 0.678,
|
| 53 |
+
"step": 3000
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.14,
|
| 57 |
+
"learning_rate": 4.644006681877445e-05,
|
| 58 |
+
"loss": 0.6673,
|
| 59 |
+
"step": 3500
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.16,
|
| 63 |
+
"learning_rate": 4.593077737940026e-05,
|
| 64 |
+
"loss": 0.6418,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.16,
|
| 69 |
+
"eval_accuracy": 0.7413654618473896,
|
| 70 |
+
"eval_loss": 0.6337006092071533,
|
| 71 |
+
"eval_runtime": 1.3627,
|
| 72 |
+
"eval_samples_per_second": 1827.229,
|
| 73 |
+
"eval_steps_per_second": 114.477,
|
| 74 |
+
"step": 4000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.18,
|
| 78 |
+
"learning_rate": 4.5421487940026075e-05,
|
| 79 |
+
"loss": 0.6562,
|
| 80 |
+
"step": 4500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.2,
|
| 84 |
+
"learning_rate": 4.491219850065189e-05,
|
| 85 |
+
"loss": 0.6461,
|
| 86 |
+
"step": 5000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.22,
|
| 90 |
+
"learning_rate": 4.440290906127771e-05,
|
| 91 |
+
"loss": 0.648,
|
| 92 |
+
"step": 5500
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.24,
|
| 96 |
+
"learning_rate": 4.389361962190352e-05,
|
| 97 |
+
"loss": 0.6207,
|
| 98 |
+
"step": 6000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.24,
|
| 102 |
+
"eval_accuracy": 0.7421686746987952,
|
| 103 |
+
"eval_loss": 0.6351802945137024,
|
| 104 |
+
"eval_runtime": 3.0952,
|
| 105 |
+
"eval_samples_per_second": 804.479,
|
| 106 |
+
"eval_steps_per_second": 50.401,
|
| 107 |
+
"step": 6000
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.26,
|
| 111 |
+
"learning_rate": 4.3385348761408085e-05,
|
| 112 |
+
"loss": 0.6335,
|
| 113 |
+
"step": 6500
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.29,
|
| 117 |
+
"learning_rate": 4.28760593220339e-05,
|
| 118 |
+
"loss": 0.633,
|
| 119 |
+
"step": 7000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.31,
|
| 123 |
+
"learning_rate": 4.236676988265971e-05,
|
| 124 |
+
"loss": 0.6173,
|
| 125 |
+
"step": 7500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.33,
|
| 129 |
+
"learning_rate": 4.185849902216428e-05,
|
| 130 |
+
"loss": 0.6146,
|
| 131 |
+
"step": 8000
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.33,
|
| 135 |
+
"eval_accuracy": 0.7534136546184739,
|
| 136 |
+
"eval_loss": 0.6005685329437256,
|
| 137 |
+
"eval_runtime": 1.3569,
|
| 138 |
+
"eval_samples_per_second": 1835.093,
|
| 139 |
+
"eval_steps_per_second": 114.97,
|
| 140 |
+
"step": 8000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.35,
|
| 144 |
+
"learning_rate": 4.1349209582790094e-05,
|
| 145 |
+
"loss": 0.6117,
|
| 146 |
+
"step": 8500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.37,
|
| 150 |
+
"learning_rate": 4.083992014341591e-05,
|
| 151 |
+
"loss": 0.6029,
|
| 152 |
+
"step": 9000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.39,
|
| 156 |
+
"learning_rate": 4.033063070404172e-05,
|
| 157 |
+
"loss": 0.607,
|
| 158 |
+
"step": 9500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.41,
|
| 162 |
+
"learning_rate": 3.9821341264667536e-05,
|
| 163 |
+
"loss": 0.593,
|
| 164 |
+
"step": 10000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.41,
|
| 168 |
+
"eval_accuracy": 0.7706827309236948,
|
| 169 |
+
"eval_loss": 0.5669934749603271,
|
| 170 |
+
"eval_runtime": 1.3711,
|
| 171 |
+
"eval_samples_per_second": 1816.031,
|
| 172 |
+
"eval_steps_per_second": 113.775,
|
| 173 |
+
"step": 10000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.43,
|
| 177 |
+
"learning_rate": 3.931205182529335e-05,
|
| 178 |
+
"loss": 0.5915,
|
| 179 |
+
"step": 10500
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.45,
|
| 183 |
+
"learning_rate": 3.8802762385919165e-05,
|
| 184 |
+
"loss": 0.593,
|
| 185 |
+
"step": 11000
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.47,
|
| 189 |
+
"learning_rate": 3.8293472946544985e-05,
|
| 190 |
+
"loss": 0.5951,
|
| 191 |
+
"step": 11500
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.49,
|
| 195 |
+
"learning_rate": 3.778418350717079e-05,
|
| 196 |
+
"loss": 0.5794,
|
| 197 |
+
"step": 12000
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 0.49,
|
| 201 |
+
"eval_accuracy": 0.770281124497992,
|
| 202 |
+
"eval_loss": 0.5802505016326904,
|
| 203 |
+
"eval_runtime": 3.0751,
|
| 204 |
+
"eval_samples_per_second": 809.728,
|
| 205 |
+
"eval_steps_per_second": 50.73,
|
| 206 |
+
"step": 12000
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.51,
|
| 210 |
+
"learning_rate": 3.7274894067796613e-05,
|
| 211 |
+
"loss": 0.5824,
|
| 212 |
+
"step": 12500
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.53,
|
| 216 |
+
"learning_rate": 3.676560462842243e-05,
|
| 217 |
+
"loss": 0.6039,
|
| 218 |
+
"step": 13000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.55,
|
| 222 |
+
"learning_rate": 3.625631518904824e-05,
|
| 223 |
+
"loss": 0.5873,
|
| 224 |
+
"step": 13500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.57,
|
| 228 |
+
"learning_rate": 3.5747025749674056e-05,
|
| 229 |
+
"loss": 0.5857,
|
| 230 |
+
"step": 14000
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.57,
|
| 234 |
+
"eval_accuracy": 0.7690763052208835,
|
| 235 |
+
"eval_loss": 0.5731454491615295,
|
| 236 |
+
"eval_runtime": 1.374,
|
| 237 |
+
"eval_samples_per_second": 1812.235,
|
| 238 |
+
"eval_steps_per_second": 113.538,
|
| 239 |
+
"step": 14000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.59,
|
| 243 |
+
"learning_rate": 3.5237736310299876e-05,
|
| 244 |
+
"loss": 0.5673,
|
| 245 |
+
"step": 14500
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 0.61,
|
| 249 |
+
"learning_rate": 3.4728446870925684e-05,
|
| 250 |
+
"loss": 0.5966,
|
| 251 |
+
"step": 15000
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 0.63,
|
| 255 |
+
"learning_rate": 3.422017601043025e-05,
|
| 256 |
+
"loss": 0.5709,
|
| 257 |
+
"step": 15500
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.65,
|
| 261 |
+
"learning_rate": 3.371190514993481e-05,
|
| 262 |
+
"loss": 0.573,
|
| 263 |
+
"step": 16000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 0.65,
|
| 267 |
+
"eval_accuracy": 0.791566265060241,
|
| 268 |
+
"eval_loss": 0.5376591086387634,
|
| 269 |
+
"eval_runtime": 1.3604,
|
| 270 |
+
"eval_samples_per_second": 1830.288,
|
| 271 |
+
"eval_steps_per_second": 114.669,
|
| 272 |
+
"step": 16000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.67,
|
| 276 |
+
"learning_rate": 3.3202615710560626e-05,
|
| 277 |
+
"loss": 0.569,
|
| 278 |
+
"step": 16500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.69,
|
| 282 |
+
"learning_rate": 3.2694344850065186e-05,
|
| 283 |
+
"loss": 0.5577,
|
| 284 |
+
"step": 17000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.71,
|
| 288 |
+
"learning_rate": 3.218505541069101e-05,
|
| 289 |
+
"loss": 0.5709,
|
| 290 |
+
"step": 17500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.73,
|
| 294 |
+
"learning_rate": 3.1675765971316814e-05,
|
| 295 |
+
"loss": 0.5475,
|
| 296 |
+
"step": 18000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.73,
|
| 300 |
+
"eval_accuracy": 0.7803212851405622,
|
| 301 |
+
"eval_loss": 0.5562066435813904,
|
| 302 |
+
"eval_runtime": 1.3712,
|
| 303 |
+
"eval_samples_per_second": 1815.926,
|
| 304 |
+
"eval_steps_per_second": 113.769,
|
| 305 |
+
"step": 18000
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.75,
|
| 309 |
+
"learning_rate": 3.1166476531942635e-05,
|
| 310 |
+
"loss": 0.5466,
|
| 311 |
+
"step": 18500
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.77,
|
| 315 |
+
"learning_rate": 3.065718709256845e-05,
|
| 316 |
+
"loss": 0.5616,
|
| 317 |
+
"step": 19000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.79,
|
| 321 |
+
"learning_rate": 3.0147897653194263e-05,
|
| 322 |
+
"loss": 0.563,
|
| 323 |
+
"step": 19500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 0.81,
|
| 327 |
+
"learning_rate": 2.963860821382008e-05,
|
| 328 |
+
"loss": 0.5438,
|
| 329 |
+
"step": 20000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.81,
|
| 333 |
+
"eval_accuracy": 0.7843373493975904,
|
| 334 |
+
"eval_loss": 0.533427894115448,
|
| 335 |
+
"eval_runtime": 3.0567,
|
| 336 |
+
"eval_samples_per_second": 814.606,
|
| 337 |
+
"eval_steps_per_second": 51.036,
|
| 338 |
+
"step": 20000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.84,
|
| 342 |
+
"learning_rate": 2.9129318774445895e-05,
|
| 343 |
+
"loss": 0.5547,
|
| 344 |
+
"step": 20500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.86,
|
| 348 |
+
"learning_rate": 2.862002933507171e-05,
|
| 349 |
+
"loss": 0.5488,
|
| 350 |
+
"step": 21000
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.88,
|
| 354 |
+
"learning_rate": 2.8110739895697523e-05,
|
| 355 |
+
"loss": 0.5357,
|
| 356 |
+
"step": 21500
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.9,
|
| 360 |
+
"learning_rate": 2.760145045632334e-05,
|
| 361 |
+
"loss": 0.5522,
|
| 362 |
+
"step": 22000
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.9,
|
| 366 |
+
"eval_accuracy": 0.7927710843373494,
|
| 367 |
+
"eval_loss": 0.5198507905006409,
|
| 368 |
+
"eval_runtime": 1.361,
|
| 369 |
+
"eval_samples_per_second": 1829.583,
|
| 370 |
+
"eval_steps_per_second": 114.624,
|
| 371 |
+
"step": 22000
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 0.92,
|
| 375 |
+
"learning_rate": 2.709216101694915e-05,
|
| 376 |
+
"loss": 0.5422,
|
| 377 |
+
"step": 22500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.94,
|
| 381 |
+
"learning_rate": 2.6582871577574968e-05,
|
| 382 |
+
"loss": 0.5481,
|
| 383 |
+
"step": 23000
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 0.96,
|
| 387 |
+
"learning_rate": 2.6073582138200786e-05,
|
| 388 |
+
"loss": 0.5376,
|
| 389 |
+
"step": 23500
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.98,
|
| 393 |
+
"learning_rate": 2.5564292698826596e-05,
|
| 394 |
+
"loss": 0.5401,
|
| 395 |
+
"step": 24000
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.98,
|
| 399 |
+
"eval_accuracy": 0.7799196787148595,
|
| 400 |
+
"eval_loss": 0.5350679159164429,
|
| 401 |
+
"eval_runtime": 3.1079,
|
| 402 |
+
"eval_samples_per_second": 801.173,
|
| 403 |
+
"eval_steps_per_second": 50.194,
|
| 404 |
+
"step": 24000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 1.0,
|
| 408 |
+
"learning_rate": 2.5055003259452414e-05,
|
| 409 |
+
"loss": 0.5375,
|
| 410 |
+
"step": 24500
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 1.02,
|
| 414 |
+
"learning_rate": 2.4545713820078228e-05,
|
| 415 |
+
"loss": 0.4512,
|
| 416 |
+
"step": 25000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.04,
|
| 420 |
+
"learning_rate": 2.403744295958279e-05,
|
| 421 |
+
"loss": 0.4575,
|
| 422 |
+
"step": 25500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.06,
|
| 426 |
+
"learning_rate": 2.3528153520208606e-05,
|
| 427 |
+
"loss": 0.4434,
|
| 428 |
+
"step": 26000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 1.06,
|
| 432 |
+
"eval_accuracy": 0.797991967871486,
|
| 433 |
+
"eval_loss": 0.5054460763931274,
|
| 434 |
+
"eval_runtime": 1.3491,
|
| 435 |
+
"eval_samples_per_second": 1845.65,
|
| 436 |
+
"eval_steps_per_second": 115.631,
|
| 437 |
+
"step": 26000
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.08,
|
| 441 |
+
"learning_rate": 2.301886408083442e-05,
|
| 442 |
+
"loss": 0.4493,
|
| 443 |
+
"step": 26500
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.1,
|
| 447 |
+
"learning_rate": 2.2509574641460234e-05,
|
| 448 |
+
"loss": 0.4281,
|
| 449 |
+
"step": 27000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 1.12,
|
| 453 |
+
"learning_rate": 2.2001303780964798e-05,
|
| 454 |
+
"loss": 0.4491,
|
| 455 |
+
"step": 27500
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 1.14,
|
| 459 |
+
"learning_rate": 2.1492014341590612e-05,
|
| 460 |
+
"loss": 0.4483,
|
| 461 |
+
"step": 28000
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 1.14,
|
| 465 |
+
"eval_accuracy": 0.7959839357429719,
|
| 466 |
+
"eval_loss": 0.5253944396972656,
|
| 467 |
+
"eval_runtime": 1.3702,
|
| 468 |
+
"eval_samples_per_second": 1817.309,
|
| 469 |
+
"eval_steps_per_second": 113.856,
|
| 470 |
+
"step": 28000
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 1.16,
|
| 474 |
+
"learning_rate": 2.098374348109518e-05,
|
| 475 |
+
"loss": 0.4543,
|
| 476 |
+
"step": 28500
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 1.18,
|
| 480 |
+
"learning_rate": 2.0474454041720993e-05,
|
| 481 |
+
"loss": 0.4405,
|
| 482 |
+
"step": 29000
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 1.2,
|
| 486 |
+
"learning_rate": 1.9965164602346807e-05,
|
| 487 |
+
"loss": 0.4563,
|
| 488 |
+
"step": 29500
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 1.22,
|
| 492 |
+
"learning_rate": 1.945587516297262e-05,
|
| 493 |
+
"loss": 0.4488,
|
| 494 |
+
"step": 30000
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 1.22,
|
| 498 |
+
"eval_accuracy": 0.7995983935742972,
|
| 499 |
+
"eval_loss": 0.5188587307929993,
|
| 500 |
+
"eval_runtime": 1.3722,
|
| 501 |
+
"eval_samples_per_second": 1814.556,
|
| 502 |
+
"eval_steps_per_second": 113.683,
|
| 503 |
+
"step": 30000
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 1.24,
|
| 507 |
+
"learning_rate": 1.8946585723598436e-05,
|
| 508 |
+
"loss": 0.4416,
|
| 509 |
+
"step": 30500
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 1.26,
|
| 513 |
+
"learning_rate": 1.843729628422425e-05,
|
| 514 |
+
"loss": 0.4289,
|
| 515 |
+
"step": 31000
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 1.28,
|
| 519 |
+
"learning_rate": 1.7928006844850064e-05,
|
| 520 |
+
"loss": 0.4315,
|
| 521 |
+
"step": 31500
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 1.3,
|
| 525 |
+
"learning_rate": 1.741871740547588e-05,
|
| 526 |
+
"loss": 0.4348,
|
| 527 |
+
"step": 32000
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.3,
|
| 531 |
+
"eval_accuracy": 0.8008032128514057,
|
| 532 |
+
"eval_loss": 0.5141127705574036,
|
| 533 |
+
"eval_runtime": 1.3695,
|
| 534 |
+
"eval_samples_per_second": 1818.24,
|
| 535 |
+
"eval_steps_per_second": 113.914,
|
| 536 |
+
"step": 32000
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 1.32,
|
| 540 |
+
"learning_rate": 1.6909427966101695e-05,
|
| 541 |
+
"loss": 0.4364,
|
| 542 |
+
"step": 32500
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.34,
|
| 546 |
+
"learning_rate": 1.640115710560626e-05,
|
| 547 |
+
"loss": 0.4398,
|
| 548 |
+
"step": 33000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.36,
|
| 552 |
+
"learning_rate": 1.5891867666232073e-05,
|
| 553 |
+
"loss": 0.4492,
|
| 554 |
+
"step": 33500
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 1.39,
|
| 558 |
+
"learning_rate": 1.5382578226857887e-05,
|
| 559 |
+
"loss": 0.4339,
|
| 560 |
+
"step": 34000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 1.39,
|
| 564 |
+
"eval_accuracy": 0.802008032128514,
|
| 565 |
+
"eval_loss": 0.5085433721542358,
|
| 566 |
+
"eval_runtime": 3.0786,
|
| 567 |
+
"eval_samples_per_second": 808.808,
|
| 568 |
+
"eval_steps_per_second": 50.672,
|
| 569 |
+
"step": 34000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.41,
|
| 573 |
+
"learning_rate": 1.4873288787483703e-05,
|
| 574 |
+
"loss": 0.4313,
|
| 575 |
+
"step": 34500
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 1.43,
|
| 579 |
+
"learning_rate": 1.4363999348109519e-05,
|
| 580 |
+
"loss": 0.4334,
|
| 581 |
+
"step": 35000
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 1.45,
|
| 585 |
+
"learning_rate": 1.3854709908735333e-05,
|
| 586 |
+
"loss": 0.4294,
|
| 587 |
+
"step": 35500
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 1.47,
|
| 591 |
+
"learning_rate": 1.3345420469361147e-05,
|
| 592 |
+
"loss": 0.4227,
|
| 593 |
+
"step": 36000
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 1.47,
|
| 597 |
+
"eval_accuracy": 0.7943775100401607,
|
| 598 |
+
"eval_loss": 0.5210641622543335,
|
| 599 |
+
"eval_runtime": 1.3741,
|
| 600 |
+
"eval_samples_per_second": 1812.132,
|
| 601 |
+
"eval_steps_per_second": 113.531,
|
| 602 |
+
"step": 36000
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 1.49,
|
| 606 |
+
"learning_rate": 1.2837149608865712e-05,
|
| 607 |
+
"loss": 0.4224,
|
| 608 |
+
"step": 36500
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 1.51,
|
| 612 |
+
"learning_rate": 1.2327860169491526e-05,
|
| 613 |
+
"loss": 0.4398,
|
| 614 |
+
"step": 37000
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 1.53,
|
| 618 |
+
"learning_rate": 1.1819589308996089e-05,
|
| 619 |
+
"loss": 0.4264,
|
| 620 |
+
"step": 37500
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 1.55,
|
| 624 |
+
"learning_rate": 1.1310299869621905e-05,
|
| 625 |
+
"loss": 0.4379,
|
| 626 |
+
"step": 38000
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 1.55,
|
| 630 |
+
"eval_accuracy": 0.802008032128514,
|
| 631 |
+
"eval_loss": 0.5011274814605713,
|
| 632 |
+
"eval_runtime": 3.0414,
|
| 633 |
+
"eval_samples_per_second": 818.69,
|
| 634 |
+
"eval_steps_per_second": 51.291,
|
| 635 |
+
"step": 38000
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 1.57,
|
| 639 |
+
"learning_rate": 1.0801010430247719e-05,
|
| 640 |
+
"loss": 0.4328,
|
| 641 |
+
"step": 38500
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 1.59,
|
| 645 |
+
"learning_rate": 1.0291720990873534e-05,
|
| 646 |
+
"loss": 0.4166,
|
| 647 |
+
"step": 39000
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.61,
|
| 651 |
+
"learning_rate": 9.78243155149935e-06,
|
| 652 |
+
"loss": 0.4294,
|
| 653 |
+
"step": 39500
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.63,
|
| 657 |
+
"learning_rate": 9.273142112125164e-06,
|
| 658 |
+
"loss": 0.4301,
|
| 659 |
+
"step": 40000
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 1.63,
|
| 663 |
+
"eval_accuracy": 0.802008032128514,
|
| 664 |
+
"eval_loss": 0.5149931311607361,
|
| 665 |
+
"eval_runtime": 1.375,
|
| 666 |
+
"eval_samples_per_second": 1810.901,
|
| 667 |
+
"eval_steps_per_second": 113.454,
|
| 668 |
+
"step": 40000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.65,
|
| 672 |
+
"learning_rate": 8.763852672750978e-06,
|
| 673 |
+
"loss": 0.4199,
|
| 674 |
+
"step": 40500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.67,
|
| 678 |
+
"learning_rate": 8.255581812255542e-06,
|
| 679 |
+
"loss": 0.43,
|
| 680 |
+
"step": 41000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 1.69,
|
| 684 |
+
"learning_rate": 7.747310951760104e-06,
|
| 685 |
+
"loss": 0.4209,
|
| 686 |
+
"step": 41500
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 1.71,
|
| 690 |
+
"learning_rate": 7.23802151238592e-06,
|
| 691 |
+
"loss": 0.4235,
|
| 692 |
+
"step": 42000
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 1.71,
|
| 696 |
+
"eval_accuracy": 0.8028112449799196,
|
| 697 |
+
"eval_loss": 0.51395583152771,
|
| 698 |
+
"eval_runtime": 1.3678,
|
| 699 |
+
"eval_samples_per_second": 1820.447,
|
| 700 |
+
"eval_steps_per_second": 114.052,
|
| 701 |
+
"step": 42000
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 1.73,
|
| 705 |
+
"learning_rate": 6.728732073011734e-06,
|
| 706 |
+
"loss": 0.4249,
|
| 707 |
+
"step": 42500
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 1.75,
|
| 711 |
+
"learning_rate": 6.219442633637549e-06,
|
| 712 |
+
"loss": 0.4102,
|
| 713 |
+
"step": 43000
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 1.77,
|
| 717 |
+
"learning_rate": 5.710153194263364e-06,
|
| 718 |
+
"loss": 0.4177,
|
| 719 |
+
"step": 43500
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 1.79,
|
| 723 |
+
"learning_rate": 5.200863754889179e-06,
|
| 724 |
+
"loss": 0.4122,
|
| 725 |
+
"step": 44000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 1.79,
|
| 729 |
+
"eval_accuracy": 0.8068273092369478,
|
| 730 |
+
"eval_loss": 0.5007363557815552,
|
| 731 |
+
"eval_runtime": 1.3698,
|
| 732 |
+
"eval_samples_per_second": 1817.792,
|
| 733 |
+
"eval_steps_per_second": 113.886,
|
| 734 |
+
"step": 44000
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 1.81,
|
| 738 |
+
"learning_rate": 4.691574315514994e-06,
|
| 739 |
+
"loss": 0.4256,
|
| 740 |
+
"step": 44500
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 1.83,
|
| 744 |
+
"learning_rate": 4.182284876140809e-06,
|
| 745 |
+
"loss": 0.4202,
|
| 746 |
+
"step": 45000
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 1.85,
|
| 750 |
+
"learning_rate": 3.6729954367666235e-06,
|
| 751 |
+
"loss": 0.4073,
|
| 752 |
+
"step": 45500
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.87,
|
| 756 |
+
"learning_rate": 3.1637059973924383e-06,
|
| 757 |
+
"loss": 0.4105,
|
| 758 |
+
"step": 46000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.87,
|
| 762 |
+
"eval_accuracy": 0.8100401606425702,
|
| 763 |
+
"eval_loss": 0.5013556480407715,
|
| 764 |
+
"eval_runtime": 1.3597,
|
| 765 |
+
"eval_samples_per_second": 1831.353,
|
| 766 |
+
"eval_steps_per_second": 114.735,
|
| 767 |
+
"step": 46000
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 1.89,
|
| 771 |
+
"learning_rate": 2.654416558018253e-06,
|
| 772 |
+
"loss": 0.4171,
|
| 773 |
+
"step": 46500
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 1.91,
|
| 777 |
+
"learning_rate": 2.1451271186440677e-06,
|
| 778 |
+
"loss": 0.4062,
|
| 779 |
+
"step": 47000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.94,
|
| 783 |
+
"learning_rate": 1.6358376792698828e-06,
|
| 784 |
+
"loss": 0.4137,
|
| 785 |
+
"step": 47500
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 1.96,
|
| 789 |
+
"learning_rate": 1.127566818774446e-06,
|
| 790 |
+
"loss": 0.3977,
|
| 791 |
+
"step": 48000
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 1.96,
|
| 795 |
+
"eval_accuracy": 0.804417670682731,
|
| 796 |
+
"eval_loss": 0.5082036852836609,
|
| 797 |
+
"eval_runtime": 1.4228,
|
| 798 |
+
"eval_samples_per_second": 1750.022,
|
| 799 |
+
"eval_steps_per_second": 109.64,
|
| 800 |
+
"step": 48000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.98,
|
| 804 |
+
"learning_rate": 6.192959582790092e-07,
|
| 805 |
+
"loss": 0.4197,
|
| 806 |
+
"step": 48500
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 2.0,
|
| 810 |
+
"learning_rate": 1.1000651890482399e-07,
|
| 811 |
+
"loss": 0.4186,
|
| 812 |
+
"step": 49000
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 2.0,
|
| 816 |
+
"step": 49088,
|
| 817 |
+
"total_flos": 3200911484154624.0,
|
| 818 |
+
"train_loss": 0.5175694002632678,
|
| 819 |
+
"train_runtime": 2940.9235,
|
| 820 |
+
"train_samples_per_second": 267.06,
|
| 821 |
+
"train_steps_per_second": 16.691
|
| 822 |
+
}
|
| 823 |
+
],
|
| 824 |
+
"max_steps": 49088,
|
| 825 |
+
"num_train_epochs": 2,
|
| 826 |
+
"total_flos": 3200911484154624.0,
|
| 827 |
+
"trial_name": null,
|
| 828 |
+
"trial_params": null
|
| 829 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3e1f177b465482b1c706c3565ba1c2dfc895903139ad41a5e22816cbfc8b2959
|
| 3 |
+
size 2863
|