Commit
•
84a4bc2
1
Parent(s):
e98c9a0
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 +1642 -0
- training_args.bin +3 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.7963855421686747,
|
| 4 |
+
"eval_loss": 0.5178616046905518,
|
| 5 |
+
"eval_runtime": 3.1592,
|
| 6 |
+
"eval_samples": 2490,
|
| 7 |
+
"eval_samples_per_second": 788.167,
|
| 8 |
+
"eval_steps_per_second": 49.379,
|
| 9 |
+
"train_loss": 0.4772139054466879,
|
| 10 |
+
"train_runtime": 9899.244,
|
| 11 |
+
"train_samples": 392702,
|
| 12 |
+
"train_samples_per_second": 158.68,
|
| 13 |
+
"train_steps_per_second": 9.918
|
| 14 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "CenIA/albert_large_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": 1024,
|
| 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": 4096,
|
| 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": 16,
|
| 35 |
+
"num_hidden_groups": 1,
|
| 36 |
+
"num_hidden_layers": 24,
|
| 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": 4.0,
|
| 3 |
+
"eval_accuracy": 0.7963855421686747,
|
| 4 |
+
"eval_loss": 0.5178616046905518,
|
| 5 |
+
"eval_runtime": 3.1592,
|
| 6 |
+
"eval_samples": 2490,
|
| 7 |
+
"eval_samples_per_second": 788.167,
|
| 8 |
+
"eval_steps_per_second": 49.379
|
| 9 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:72b1d305b80a49080aef77b29450cd97d3382cf6358f31afdd22867c98340cef
|
| 3 |
+
size 71275217
|
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_large_spanish", "tokenizer_class": "AlbertTokenizer"}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"train_loss": 0.4772139054466879,
|
| 4 |
+
"train_runtime": 9899.244,
|
| 5 |
+
"train_samples": 392702,
|
| 6 |
+
"train_samples_per_second": 158.68,
|
| 7 |
+
"train_steps_per_second": 9.918
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1642 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.5178616046905518,
|
| 3 |
+
"best_model_checkpoint": "/data/jcanete/all_results/xnli/albeto_large/epochs_4_bs_16_lr_5e-6/checkpoint-48000",
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"global_step": 98176,
|
| 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": 4.974790172750978e-06,
|
| 13 |
+
"loss": 1.0068,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.04,
|
| 18 |
+
"learning_rate": 4.9493766297262065e-06,
|
| 19 |
+
"loss": 0.8946,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.06,
|
| 24 |
+
"learning_rate": 4.924268660365059e-06,
|
| 25 |
+
"loss": 0.8615,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.08,
|
| 30 |
+
"learning_rate": 4.899007904172099e-06,
|
| 31 |
+
"loss": 1.0693,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.08,
|
| 36 |
+
"eval_accuracy": 0.5301204819277109,
|
| 37 |
+
"eval_loss": 0.9672217965126038,
|
| 38 |
+
"eval_runtime": 3.1152,
|
| 39 |
+
"eval_samples_per_second": 799.309,
|
| 40 |
+
"eval_steps_per_second": 50.077,
|
| 41 |
+
"step": 2000
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.1,
|
| 45 |
+
"learning_rate": 4.873594361147327e-06,
|
| 46 |
+
"loss": 0.9457,
|
| 47 |
+
"step": 2500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.12,
|
| 51 |
+
"learning_rate": 4.848129889178619e-06,
|
| 52 |
+
"loss": 0.8099,
|
| 53 |
+
"step": 3000
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.14,
|
| 57 |
+
"learning_rate": 4.822665417209909e-06,
|
| 58 |
+
"loss": 0.7534,
|
| 59 |
+
"step": 3500
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.16,
|
| 63 |
+
"learning_rate": 4.7972009452412e-06,
|
| 64 |
+
"loss": 0.7074,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.16,
|
| 69 |
+
"eval_accuracy": 0.6859437751004016,
|
| 70 |
+
"eval_loss": 0.7257041931152344,
|
| 71 |
+
"eval_runtime": 3.1181,
|
| 72 |
+
"eval_samples_per_second": 798.568,
|
| 73 |
+
"eval_steps_per_second": 50.031,
|
| 74 |
+
"step": 4000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.18,
|
| 78 |
+
"learning_rate": 4.7717364732724905e-06,
|
| 79 |
+
"loss": 0.706,
|
| 80 |
+
"step": 4500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.2,
|
| 84 |
+
"learning_rate": 4.746272001303782e-06,
|
| 85 |
+
"loss": 0.6929,
|
| 86 |
+
"step": 5000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.22,
|
| 90 |
+
"learning_rate": 4.720807529335072e-06,
|
| 91 |
+
"loss": 0.6786,
|
| 92 |
+
"step": 5500
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.24,
|
| 96 |
+
"learning_rate": 4.6953430573663625e-06,
|
| 97 |
+
"loss": 0.6533,
|
| 98 |
+
"step": 6000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.24,
|
| 102 |
+
"eval_accuracy": 0.7156626506024096,
|
| 103 |
+
"eval_loss": 0.7040365934371948,
|
| 104 |
+
"eval_runtime": 3.106,
|
| 105 |
+
"eval_samples_per_second": 801.678,
|
| 106 |
+
"eval_steps_per_second": 50.226,
|
| 107 |
+
"step": 6000
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.26,
|
| 111 |
+
"learning_rate": 4.669878585397654e-06,
|
| 112 |
+
"loss": 0.7996,
|
| 113 |
+
"step": 6500
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.29,
|
| 117 |
+
"learning_rate": 4.644414113428944e-06,
|
| 118 |
+
"loss": 0.7019,
|
| 119 |
+
"step": 7000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.31,
|
| 123 |
+
"learning_rate": 4.618949641460235e-06,
|
| 124 |
+
"loss": 0.6643,
|
| 125 |
+
"step": 7500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.33,
|
| 129 |
+
"learning_rate": 4.593485169491526e-06,
|
| 130 |
+
"loss": 0.6375,
|
| 131 |
+
"step": 8000
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.33,
|
| 135 |
+
"eval_accuracy": 0.7421686746987952,
|
| 136 |
+
"eval_loss": 0.6504086256027222,
|
| 137 |
+
"eval_runtime": 3.105,
|
| 138 |
+
"eval_samples_per_second": 801.928,
|
| 139 |
+
"eval_steps_per_second": 50.241,
|
| 140 |
+
"step": 8000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.35,
|
| 144 |
+
"learning_rate": 4.568020697522816e-06,
|
| 145 |
+
"loss": 0.6397,
|
| 146 |
+
"step": 8500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.37,
|
| 150 |
+
"learning_rate": 4.542556225554107e-06,
|
| 151 |
+
"loss": 0.6264,
|
| 152 |
+
"step": 9000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.39,
|
| 156 |
+
"learning_rate": 4.517091753585398e-06,
|
| 157 |
+
"loss": 0.6288,
|
| 158 |
+
"step": 9500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.41,
|
| 162 |
+
"learning_rate": 4.491627281616689e-06,
|
| 163 |
+
"loss": 0.6158,
|
| 164 |
+
"step": 10000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.41,
|
| 168 |
+
"eval_accuracy": 0.7542168674698795,
|
| 169 |
+
"eval_loss": 0.6070215702056885,
|
| 170 |
+
"eval_runtime": 3.1074,
|
| 171 |
+
"eval_samples_per_second": 801.321,
|
| 172 |
+
"eval_steps_per_second": 50.203,
|
| 173 |
+
"step": 10000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.43,
|
| 177 |
+
"learning_rate": 4.46616280964798e-06,
|
| 178 |
+
"loss": 0.6155,
|
| 179 |
+
"step": 10500
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.45,
|
| 183 |
+
"learning_rate": 4.44069833767927e-06,
|
| 184 |
+
"loss": 0.6139,
|
| 185 |
+
"step": 11000
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.47,
|
| 189 |
+
"learning_rate": 4.415233865710561e-06,
|
| 190 |
+
"loss": 0.6096,
|
| 191 |
+
"step": 11500
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.49,
|
| 195 |
+
"learning_rate": 4.389769393741852e-06,
|
| 196 |
+
"loss": 0.5992,
|
| 197 |
+
"step": 12000
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 0.49,
|
| 201 |
+
"eval_accuracy": 0.7674698795180723,
|
| 202 |
+
"eval_loss": 0.5774316191673279,
|
| 203 |
+
"eval_runtime": 3.0887,
|
| 204 |
+
"eval_samples_per_second": 806.17,
|
| 205 |
+
"eval_steps_per_second": 50.507,
|
| 206 |
+
"step": 12000
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.51,
|
| 210 |
+
"learning_rate": 4.364304921773142e-06,
|
| 211 |
+
"loss": 0.6005,
|
| 212 |
+
"step": 12500
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.53,
|
| 216 |
+
"learning_rate": 4.3388404498044335e-06,
|
| 217 |
+
"loss": 0.6118,
|
| 218 |
+
"step": 13000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.55,
|
| 222 |
+
"learning_rate": 4.313375977835724e-06,
|
| 223 |
+
"loss": 0.6063,
|
| 224 |
+
"step": 13500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.57,
|
| 228 |
+
"learning_rate": 4.287911505867014e-06,
|
| 229 |
+
"loss": 0.5997,
|
| 230 |
+
"step": 14000
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.57,
|
| 234 |
+
"eval_accuracy": 0.7714859437751004,
|
| 235 |
+
"eval_loss": 0.5825369954109192,
|
| 236 |
+
"eval_runtime": 3.1126,
|
| 237 |
+
"eval_samples_per_second": 799.965,
|
| 238 |
+
"eval_steps_per_second": 50.118,
|
| 239 |
+
"step": 14000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.59,
|
| 243 |
+
"learning_rate": 4.2624470338983054e-06,
|
| 244 |
+
"loss": 0.5779,
|
| 245 |
+
"step": 14500
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 0.61,
|
| 249 |
+
"learning_rate": 4.236982561929597e-06,
|
| 250 |
+
"loss": 0.6107,
|
| 251 |
+
"step": 15000
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 0.63,
|
| 255 |
+
"learning_rate": 4.211518089960887e-06,
|
| 256 |
+
"loss": 0.5816,
|
| 257 |
+
"step": 15500
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.65,
|
| 261 |
+
"learning_rate": 4.186053617992178e-06,
|
| 262 |
+
"loss": 0.592,
|
| 263 |
+
"step": 16000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 0.65,
|
| 267 |
+
"eval_accuracy": 0.7742971887550201,
|
| 268 |
+
"eval_loss": 0.5646522641181946,
|
| 269 |
+
"eval_runtime": 3.1034,
|
| 270 |
+
"eval_samples_per_second": 802.352,
|
| 271 |
+
"eval_steps_per_second": 50.268,
|
| 272 |
+
"step": 16000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.67,
|
| 276 |
+
"learning_rate": 4.1605891460234686e-06,
|
| 277 |
+
"loss": 0.5846,
|
| 278 |
+
"step": 16500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.69,
|
| 282 |
+
"learning_rate": 4.135124674054759e-06,
|
| 283 |
+
"loss": 0.5791,
|
| 284 |
+
"step": 17000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.71,
|
| 288 |
+
"learning_rate": 4.10966020208605e-06,
|
| 289 |
+
"loss": 0.595,
|
| 290 |
+
"step": 17500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.73,
|
| 294 |
+
"learning_rate": 4.0841957301173405e-06,
|
| 295 |
+
"loss": 0.5682,
|
| 296 |
+
"step": 18000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.73,
|
| 300 |
+
"eval_accuracy": 0.7746987951807229,
|
| 301 |
+
"eval_loss": 0.5749292969703674,
|
| 302 |
+
"eval_runtime": 3.0899,
|
| 303 |
+
"eval_samples_per_second": 805.855,
|
| 304 |
+
"eval_steps_per_second": 50.487,
|
| 305 |
+
"step": 18000
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.75,
|
| 309 |
+
"learning_rate": 4.058731258148632e-06,
|
| 310 |
+
"loss": 0.5634,
|
| 311 |
+
"step": 18500
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.77,
|
| 315 |
+
"learning_rate": 4.033266786179922e-06,
|
| 316 |
+
"loss": 0.5767,
|
| 317 |
+
"step": 19000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.79,
|
| 321 |
+
"learning_rate": 4.0078023142112124e-06,
|
| 322 |
+
"loss": 0.5757,
|
| 323 |
+
"step": 19500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 0.81,
|
| 327 |
+
"learning_rate": 3.982337842242504e-06,
|
| 328 |
+
"loss": 0.5549,
|
| 329 |
+
"step": 20000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.81,
|
| 333 |
+
"eval_accuracy": 0.7738955823293173,
|
| 334 |
+
"eval_loss": 0.5731613039970398,
|
| 335 |
+
"eval_runtime": 3.1136,
|
| 336 |
+
"eval_samples_per_second": 799.722,
|
| 337 |
+
"eval_steps_per_second": 50.103,
|
| 338 |
+
"step": 20000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.84,
|
| 342 |
+
"learning_rate": 3.956873370273794e-06,
|
| 343 |
+
"loss": 0.5704,
|
| 344 |
+
"step": 20500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.86,
|
| 348 |
+
"learning_rate": 3.931408898305085e-06,
|
| 349 |
+
"loss": 0.5668,
|
| 350 |
+
"step": 21000
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.88,
|
| 354 |
+
"learning_rate": 3.9059444263363764e-06,
|
| 355 |
+
"loss": 0.5523,
|
| 356 |
+
"step": 21500
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.9,
|
| 360 |
+
"learning_rate": 3.880479954367666e-06,
|
| 361 |
+
"loss": 0.5651,
|
| 362 |
+
"step": 22000
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.9,
|
| 366 |
+
"eval_accuracy": 0.7887550200803213,
|
| 367 |
+
"eval_loss": 0.5458173751831055,
|
| 368 |
+
"eval_runtime": 3.1133,
|
| 369 |
+
"eval_samples_per_second": 799.785,
|
| 370 |
+
"eval_steps_per_second": 50.107,
|
| 371 |
+
"step": 22000
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 0.92,
|
| 375 |
+
"learning_rate": 3.855015482398957e-06,
|
| 376 |
+
"loss": 0.5582,
|
| 377 |
+
"step": 22500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.94,
|
| 381 |
+
"learning_rate": 3.829551010430248e-06,
|
| 382 |
+
"loss": 0.5567,
|
| 383 |
+
"step": 23000
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 0.96,
|
| 387 |
+
"learning_rate": 3.8040865384615387e-06,
|
| 388 |
+
"loss": 0.5505,
|
| 389 |
+
"step": 23500
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.98,
|
| 393 |
+
"learning_rate": 3.7786220664928295e-06,
|
| 394 |
+
"loss": 0.561,
|
| 395 |
+
"step": 24000
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.98,
|
| 399 |
+
"eval_accuracy": 0.7859437751004016,
|
| 400 |
+
"eval_loss": 0.5578823089599609,
|
| 401 |
+
"eval_runtime": 3.1003,
|
| 402 |
+
"eval_samples_per_second": 803.157,
|
| 403 |
+
"eval_steps_per_second": 50.318,
|
| 404 |
+
"step": 24000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 1.0,
|
| 408 |
+
"learning_rate": 3.75315759452412e-06,
|
| 409 |
+
"loss": 0.559,
|
| 410 |
+
"step": 24500
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 1.02,
|
| 414 |
+
"learning_rate": 3.7276931225554107e-06,
|
| 415 |
+
"loss": 0.5078,
|
| 416 |
+
"step": 25000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.04,
|
| 420 |
+
"learning_rate": 3.702228650586702e-06,
|
| 421 |
+
"loss": 0.5162,
|
| 422 |
+
"step": 25500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.06,
|
| 426 |
+
"learning_rate": 3.6767641786179927e-06,
|
| 427 |
+
"loss": 0.5033,
|
| 428 |
+
"step": 26000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 1.06,
|
| 432 |
+
"eval_accuracy": 0.7803212851405622,
|
| 433 |
+
"eval_loss": 0.5404613614082336,
|
| 434 |
+
"eval_runtime": 3.1086,
|
| 435 |
+
"eval_samples_per_second": 801.01,
|
| 436 |
+
"eval_steps_per_second": 50.184,
|
| 437 |
+
"step": 26000
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.08,
|
| 441 |
+
"learning_rate": 3.6512997066492834e-06,
|
| 442 |
+
"loss": 0.5035,
|
| 443 |
+
"step": 26500
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.1,
|
| 447 |
+
"learning_rate": 3.6258352346805742e-06,
|
| 448 |
+
"loss": 0.4876,
|
| 449 |
+
"step": 27000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 1.12,
|
| 453 |
+
"learning_rate": 3.6003707627118646e-06,
|
| 454 |
+
"loss": 0.5037,
|
| 455 |
+
"step": 27500
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 1.14,
|
| 459 |
+
"learning_rate": 3.5749062907431554e-06,
|
| 460 |
+
"loss": 0.507,
|
| 461 |
+
"step": 28000
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 1.14,
|
| 465 |
+
"eval_accuracy": 0.7867469879518072,
|
| 466 |
+
"eval_loss": 0.5373629331588745,
|
| 467 |
+
"eval_runtime": 3.1109,
|
| 468 |
+
"eval_samples_per_second": 800.404,
|
| 469 |
+
"eval_steps_per_second": 50.146,
|
| 470 |
+
"step": 28000
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 1.16,
|
| 474 |
+
"learning_rate": 3.549441818774446e-06,
|
| 475 |
+
"loss": 0.5227,
|
| 476 |
+
"step": 28500
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 1.18,
|
| 480 |
+
"learning_rate": 3.523977346805737e-06,
|
| 481 |
+
"loss": 0.5025,
|
| 482 |
+
"step": 29000
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 1.2,
|
| 486 |
+
"learning_rate": 3.4985128748370277e-06,
|
| 487 |
+
"loss": 0.5202,
|
| 488 |
+
"step": 29500
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 1.22,
|
| 492 |
+
"learning_rate": 3.4730484028683185e-06,
|
| 493 |
+
"loss": 0.5027,
|
| 494 |
+
"step": 30000
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 1.22,
|
| 498 |
+
"eval_accuracy": 0.7843373493975904,
|
| 499 |
+
"eval_loss": 0.5514963865280151,
|
| 500 |
+
"eval_runtime": 3.114,
|
| 501 |
+
"eval_samples_per_second": 799.625,
|
| 502 |
+
"eval_steps_per_second": 50.097,
|
| 503 |
+
"step": 30000
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 1.24,
|
| 507 |
+
"learning_rate": 3.4476348598435467e-06,
|
| 508 |
+
"loss": 0.5052,
|
| 509 |
+
"step": 30500
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 1.26,
|
| 513 |
+
"learning_rate": 3.4221703878748375e-06,
|
| 514 |
+
"loss": 0.4932,
|
| 515 |
+
"step": 31000
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 1.28,
|
| 519 |
+
"learning_rate": 3.3967059159061283e-06,
|
| 520 |
+
"loss": 0.4916,
|
| 521 |
+
"step": 31500
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 1.3,
|
| 525 |
+
"learning_rate": 3.3712414439374187e-06,
|
| 526 |
+
"loss": 0.5021,
|
| 527 |
+
"step": 32000
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.3,
|
| 531 |
+
"eval_accuracy": 0.793574297188755,
|
| 532 |
+
"eval_loss": 0.5289486050605774,
|
| 533 |
+
"eval_runtime": 3.1115,
|
| 534 |
+
"eval_samples_per_second": 800.255,
|
| 535 |
+
"eval_steps_per_second": 50.136,
|
| 536 |
+
"step": 32000
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 1.32,
|
| 540 |
+
"learning_rate": 3.3457769719687095e-06,
|
| 541 |
+
"loss": 0.501,
|
| 542 |
+
"step": 32500
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.34,
|
| 546 |
+
"learning_rate": 3.3203125000000002e-06,
|
| 547 |
+
"loss": 0.4921,
|
| 548 |
+
"step": 33000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.36,
|
| 552 |
+
"learning_rate": 3.294848028031291e-06,
|
| 553 |
+
"loss": 0.5051,
|
| 554 |
+
"step": 33500
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 1.39,
|
| 558 |
+
"learning_rate": 3.269383556062582e-06,
|
| 559 |
+
"loss": 0.4895,
|
| 560 |
+
"step": 34000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 1.39,
|
| 564 |
+
"eval_accuracy": 0.7923694779116466,
|
| 565 |
+
"eval_loss": 0.5383234024047852,
|
| 566 |
+
"eval_runtime": 3.1113,
|
| 567 |
+
"eval_samples_per_second": 800.298,
|
| 568 |
+
"eval_steps_per_second": 50.139,
|
| 569 |
+
"step": 34000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.41,
|
| 573 |
+
"learning_rate": 3.243919084093872e-06,
|
| 574 |
+
"loss": 0.49,
|
| 575 |
+
"step": 34500
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 1.43,
|
| 579 |
+
"learning_rate": 3.218454612125163e-06,
|
| 580 |
+
"loss": 0.5014,
|
| 581 |
+
"step": 35000
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 1.45,
|
| 585 |
+
"learning_rate": 3.1930410691003916e-06,
|
| 586 |
+
"loss": 0.4877,
|
| 587 |
+
"step": 35500
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 1.47,
|
| 591 |
+
"learning_rate": 3.167576597131682e-06,
|
| 592 |
+
"loss": 0.4817,
|
| 593 |
+
"step": 36000
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 1.47,
|
| 597 |
+
"eval_accuracy": 0.7883534136546185,
|
| 598 |
+
"eval_loss": 0.5456702709197998,
|
| 599 |
+
"eval_runtime": 3.1157,
|
| 600 |
+
"eval_samples_per_second": 799.189,
|
| 601 |
+
"eval_steps_per_second": 50.07,
|
| 602 |
+
"step": 36000
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 1.49,
|
| 606 |
+
"learning_rate": 3.1421121251629727e-06,
|
| 607 |
+
"loss": 0.4889,
|
| 608 |
+
"step": 36500
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 1.51,
|
| 612 |
+
"learning_rate": 3.1166476531942635e-06,
|
| 613 |
+
"loss": 0.4928,
|
| 614 |
+
"step": 37000
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 1.53,
|
| 618 |
+
"learning_rate": 3.0911831812255543e-06,
|
| 619 |
+
"loss": 0.492,
|
| 620 |
+
"step": 37500
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 1.55,
|
| 624 |
+
"learning_rate": 3.065718709256845e-06,
|
| 625 |
+
"loss": 0.4958,
|
| 626 |
+
"step": 38000
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 1.55,
|
| 630 |
+
"eval_accuracy": 0.7983935742971887,
|
| 631 |
+
"eval_loss": 0.5244956612586975,
|
| 632 |
+
"eval_runtime": 3.1141,
|
| 633 |
+
"eval_samples_per_second": 799.583,
|
| 634 |
+
"eval_steps_per_second": 50.094,
|
| 635 |
+
"step": 38000
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 1.57,
|
| 639 |
+
"learning_rate": 3.040254237288136e-06,
|
| 640 |
+
"loss": 0.494,
|
| 641 |
+
"step": 38500
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 1.59,
|
| 645 |
+
"learning_rate": 3.0147897653194262e-06,
|
| 646 |
+
"loss": 0.4876,
|
| 647 |
+
"step": 39000
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.61,
|
| 651 |
+
"learning_rate": 2.989376222294655e-06,
|
| 652 |
+
"loss": 0.4983,
|
| 653 |
+
"step": 39500
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.63,
|
| 657 |
+
"learning_rate": 2.9639626792698826e-06,
|
| 658 |
+
"loss": 0.4993,
|
| 659 |
+
"step": 40000
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 1.63,
|
| 663 |
+
"eval_accuracy": 0.7931726907630522,
|
| 664 |
+
"eval_loss": 0.5367791652679443,
|
| 665 |
+
"eval_runtime": 3.1071,
|
| 666 |
+
"eval_samples_per_second": 801.396,
|
| 667 |
+
"eval_steps_per_second": 50.208,
|
| 668 |
+
"step": 40000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.65,
|
| 672 |
+
"learning_rate": 2.9384982073011734e-06,
|
| 673 |
+
"loss": 0.4879,
|
| 674 |
+
"step": 40500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.67,
|
| 678 |
+
"learning_rate": 2.9130337353324646e-06,
|
| 679 |
+
"loss": 0.4942,
|
| 680 |
+
"step": 41000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 1.69,
|
| 684 |
+
"learning_rate": 2.8875692633637554e-06,
|
| 685 |
+
"loss": 0.4895,
|
| 686 |
+
"step": 41500
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 1.71,
|
| 690 |
+
"learning_rate": 2.862104791395046e-06,
|
| 691 |
+
"loss": 0.4861,
|
| 692 |
+
"step": 42000
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 1.71,
|
| 696 |
+
"eval_accuracy": 0.7967871485943775,
|
| 697 |
+
"eval_loss": 0.5284856557846069,
|
| 698 |
+
"eval_runtime": 3.0957,
|
| 699 |
+
"eval_samples_per_second": 804.354,
|
| 700 |
+
"eval_steps_per_second": 50.393,
|
| 701 |
+
"step": 42000
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 1.73,
|
| 705 |
+
"learning_rate": 2.8366403194263366e-06,
|
| 706 |
+
"loss": 0.4936,
|
| 707 |
+
"step": 42500
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 1.75,
|
| 711 |
+
"learning_rate": 2.8111758474576274e-06,
|
| 712 |
+
"loss": 0.4782,
|
| 713 |
+
"step": 43000
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 1.77,
|
| 717 |
+
"learning_rate": 2.785711375488918e-06,
|
| 718 |
+
"loss": 0.4857,
|
| 719 |
+
"step": 43500
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 1.79,
|
| 723 |
+
"learning_rate": 2.760297832464146e-06,
|
| 724 |
+
"loss": 0.4865,
|
| 725 |
+
"step": 44000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 1.79,
|
| 729 |
+
"eval_accuracy": 0.7967871485943775,
|
| 730 |
+
"eval_loss": 0.5284005403518677,
|
| 731 |
+
"eval_runtime": 3.1113,
|
| 732 |
+
"eval_samples_per_second": 800.304,
|
| 733 |
+
"eval_steps_per_second": 50.14,
|
| 734 |
+
"step": 44000
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 1.81,
|
| 738 |
+
"learning_rate": 2.7348333604954367e-06,
|
| 739 |
+
"loss": 0.4913,
|
| 740 |
+
"step": 44500
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 1.83,
|
| 744 |
+
"learning_rate": 2.709368888526728e-06,
|
| 745 |
+
"loss": 0.4899,
|
| 746 |
+
"step": 45000
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 1.85,
|
| 750 |
+
"learning_rate": 2.6839044165580187e-06,
|
| 751 |
+
"loss": 0.4872,
|
| 752 |
+
"step": 45500
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.87,
|
| 756 |
+
"learning_rate": 2.6584399445893095e-06,
|
| 757 |
+
"loss": 0.4827,
|
| 758 |
+
"step": 46000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.87,
|
| 762 |
+
"eval_accuracy": 0.7927710843373494,
|
| 763 |
+
"eval_loss": 0.5268122553825378,
|
| 764 |
+
"eval_runtime": 3.1094,
|
| 765 |
+
"eval_samples_per_second": 800.807,
|
| 766 |
+
"eval_steps_per_second": 50.171,
|
| 767 |
+
"step": 46000
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 1.89,
|
| 771 |
+
"learning_rate": 2.6330264015645373e-06,
|
| 772 |
+
"loss": 0.4844,
|
| 773 |
+
"step": 46500
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 1.91,
|
| 777 |
+
"learning_rate": 2.607561929595828e-06,
|
| 778 |
+
"loss": 0.4803,
|
| 779 |
+
"step": 47000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.94,
|
| 783 |
+
"learning_rate": 2.582097457627119e-06,
|
| 784 |
+
"loss": 0.4863,
|
| 785 |
+
"step": 47500
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 1.96,
|
| 789 |
+
"learning_rate": 2.556632985658409e-06,
|
| 790 |
+
"loss": 0.4706,
|
| 791 |
+
"step": 48000
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 1.96,
|
| 795 |
+
"eval_accuracy": 0.7963855421686747,
|
| 796 |
+
"eval_loss": 0.5178616046905518,
|
| 797 |
+
"eval_runtime": 3.1089,
|
| 798 |
+
"eval_samples_per_second": 800.919,
|
| 799 |
+
"eval_steps_per_second": 50.178,
|
| 800 |
+
"step": 48000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.98,
|
| 804 |
+
"learning_rate": 2.5311685136897e-06,
|
| 805 |
+
"loss": 0.4921,
|
| 806 |
+
"step": 48500
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 2.0,
|
| 810 |
+
"learning_rate": 2.5057549706649286e-06,
|
| 811 |
+
"loss": 0.4877,
|
| 812 |
+
"step": 49000
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 2.02,
|
| 816 |
+
"learning_rate": 2.4802904986962194e-06,
|
| 817 |
+
"loss": 0.4357,
|
| 818 |
+
"step": 49500
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 2.04,
|
| 822 |
+
"learning_rate": 2.45482602672751e-06,
|
| 823 |
+
"loss": 0.4374,
|
| 824 |
+
"step": 50000
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 2.04,
|
| 828 |
+
"eval_accuracy": 0.7995983935742972,
|
| 829 |
+
"eval_loss": 0.5278629064559937,
|
| 830 |
+
"eval_runtime": 3.1508,
|
| 831 |
+
"eval_samples_per_second": 790.269,
|
| 832 |
+
"eval_steps_per_second": 49.511,
|
| 833 |
+
"step": 50000
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 2.06,
|
| 837 |
+
"learning_rate": 2.4293615547588006e-06,
|
| 838 |
+
"loss": 0.418,
|
| 839 |
+
"step": 50500
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 2.08,
|
| 843 |
+
"learning_rate": 2.4038970827900913e-06,
|
| 844 |
+
"loss": 0.4102,
|
| 845 |
+
"step": 51000
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 2.1,
|
| 849 |
+
"learning_rate": 2.3784835397653195e-06,
|
| 850 |
+
"loss": 0.4268,
|
| 851 |
+
"step": 51500
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 2.12,
|
| 855 |
+
"learning_rate": 2.3530190677966103e-06,
|
| 856 |
+
"loss": 0.4238,
|
| 857 |
+
"step": 52000
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 2.12,
|
| 861 |
+
"eval_accuracy": 0.7911646586345381,
|
| 862 |
+
"eval_loss": 0.5376625061035156,
|
| 863 |
+
"eval_runtime": 3.1036,
|
| 864 |
+
"eval_samples_per_second": 802.307,
|
| 865 |
+
"eval_steps_per_second": 50.265,
|
| 866 |
+
"step": 52000
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 2.14,
|
| 870 |
+
"learning_rate": 2.327554595827901e-06,
|
| 871 |
+
"loss": 0.419,
|
| 872 |
+
"step": 52500
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 2.16,
|
| 876 |
+
"learning_rate": 2.302090123859192e-06,
|
| 877 |
+
"loss": 0.4006,
|
| 878 |
+
"step": 53000
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 2.18,
|
| 882 |
+
"learning_rate": 2.2766765808344197e-06,
|
| 883 |
+
"loss": 0.4142,
|
| 884 |
+
"step": 53500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 2.2,
|
| 888 |
+
"learning_rate": 2.251212108865711e-06,
|
| 889 |
+
"loss": 0.4128,
|
| 890 |
+
"step": 54000
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 2.2,
|
| 894 |
+
"eval_accuracy": 0.7971887550200804,
|
| 895 |
+
"eval_loss": 0.5341054797172546,
|
| 896 |
+
"eval_runtime": 3.1779,
|
| 897 |
+
"eval_samples_per_second": 783.533,
|
| 898 |
+
"eval_steps_per_second": 49.089,
|
| 899 |
+
"step": 54000
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 2.22,
|
| 903 |
+
"learning_rate": 2.2257476368970017e-06,
|
| 904 |
+
"loss": 0.4169,
|
| 905 |
+
"step": 54500
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 2.24,
|
| 909 |
+
"learning_rate": 2.200283164928292e-06,
|
| 910 |
+
"loss": 0.4219,
|
| 911 |
+
"step": 55000
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 2.26,
|
| 915 |
+
"learning_rate": 2.174818692959583e-06,
|
| 916 |
+
"loss": 0.3998,
|
| 917 |
+
"step": 55500
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 2.28,
|
| 921 |
+
"learning_rate": 2.149405149934811e-06,
|
| 922 |
+
"loss": 0.4079,
|
| 923 |
+
"step": 56000
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 2.28,
|
| 927 |
+
"eval_accuracy": 0.8036144578313253,
|
| 928 |
+
"eval_loss": 0.53012615442276,
|
| 929 |
+
"eval_runtime": 3.1161,
|
| 930 |
+
"eval_samples_per_second": 799.075,
|
| 931 |
+
"eval_steps_per_second": 50.063,
|
| 932 |
+
"step": 56000
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 2.3,
|
| 936 |
+
"learning_rate": 2.123940677966102e-06,
|
| 937 |
+
"loss": 0.4139,
|
| 938 |
+
"step": 56500
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 2.32,
|
| 942 |
+
"learning_rate": 2.0984762059973926e-06,
|
| 943 |
+
"loss": 0.4117,
|
| 944 |
+
"step": 57000
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 2.34,
|
| 948 |
+
"learning_rate": 2.0730117340286834e-06,
|
| 949 |
+
"loss": 0.4144,
|
| 950 |
+
"step": 57500
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 2.36,
|
| 954 |
+
"learning_rate": 2.0475981910039116e-06,
|
| 955 |
+
"loss": 0.4219,
|
| 956 |
+
"step": 58000
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 2.36,
|
| 960 |
+
"eval_accuracy": 0.8,
|
| 961 |
+
"eval_loss": 0.5489892363548279,
|
| 962 |
+
"eval_runtime": 3.1097,
|
| 963 |
+
"eval_samples_per_second": 800.724,
|
| 964 |
+
"eval_steps_per_second": 50.166,
|
| 965 |
+
"step": 58000
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 2.38,
|
| 969 |
+
"learning_rate": 2.0221337190352024e-06,
|
| 970 |
+
"loss": 0.4231,
|
| 971 |
+
"step": 58500
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 2.4,
|
| 975 |
+
"learning_rate": 1.996669247066493e-06,
|
| 976 |
+
"loss": 0.4054,
|
| 977 |
+
"step": 59000
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 2.42,
|
| 981 |
+
"learning_rate": 1.9712047750977835e-06,
|
| 982 |
+
"loss": 0.4166,
|
| 983 |
+
"step": 59500
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 2.44,
|
| 987 |
+
"learning_rate": 1.9457912320730117e-06,
|
| 988 |
+
"loss": 0.3995,
|
| 989 |
+
"step": 60000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 2.44,
|
| 993 |
+
"eval_accuracy": 0.8028112449799196,
|
| 994 |
+
"eval_loss": 0.5354592204093933,
|
| 995 |
+
"eval_runtime": 3.1379,
|
| 996 |
+
"eval_samples_per_second": 793.526,
|
| 997 |
+
"eval_steps_per_second": 49.715,
|
| 998 |
+
"step": 60000
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 2.46,
|
| 1002 |
+
"learning_rate": 1.9203267601043025e-06,
|
| 1003 |
+
"loss": 0.4244,
|
| 1004 |
+
"step": 60500
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 2.49,
|
| 1008 |
+
"learning_rate": 1.8948622881355933e-06,
|
| 1009 |
+
"loss": 0.412,
|
| 1010 |
+
"step": 61000
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 2.51,
|
| 1014 |
+
"learning_rate": 1.869397816166884e-06,
|
| 1015 |
+
"loss": 0.4108,
|
| 1016 |
+
"step": 61500
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 2.53,
|
| 1020 |
+
"learning_rate": 1.8439333441981749e-06,
|
| 1021 |
+
"loss": 0.4181,
|
| 1022 |
+
"step": 62000
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 2.53,
|
| 1026 |
+
"eval_accuracy": 0.802008032128514,
|
| 1027 |
+
"eval_loss": 0.5300112366676331,
|
| 1028 |
+
"eval_runtime": 3.1096,
|
| 1029 |
+
"eval_samples_per_second": 800.752,
|
| 1030 |
+
"eval_steps_per_second": 50.168,
|
| 1031 |
+
"step": 62000
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 2.55,
|
| 1035 |
+
"learning_rate": 1.8184688722294654e-06,
|
| 1036 |
+
"loss": 0.4081,
|
| 1037 |
+
"step": 62500
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 2.57,
|
| 1041 |
+
"learning_rate": 1.7930044002607562e-06,
|
| 1042 |
+
"loss": 0.4048,
|
| 1043 |
+
"step": 63000
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 2.59,
|
| 1047 |
+
"learning_rate": 1.7675399282920472e-06,
|
| 1048 |
+
"loss": 0.429,
|
| 1049 |
+
"step": 63500
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 2.61,
|
| 1053 |
+
"learning_rate": 1.7420754563233378e-06,
|
| 1054 |
+
"loss": 0.4179,
|
| 1055 |
+
"step": 64000
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 2.61,
|
| 1059 |
+
"eval_accuracy": 0.7963855421686747,
|
| 1060 |
+
"eval_loss": 0.5444837212562561,
|
| 1061 |
+
"eval_runtime": 3.1167,
|
| 1062 |
+
"eval_samples_per_second": 798.922,
|
| 1063 |
+
"eval_steps_per_second": 50.053,
|
| 1064 |
+
"step": 64000
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 2.63,
|
| 1068 |
+
"learning_rate": 1.7166109843546286e-06,
|
| 1069 |
+
"loss": 0.4114,
|
| 1070 |
+
"step": 64500
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 2.65,
|
| 1074 |
+
"learning_rate": 1.6911465123859194e-06,
|
| 1075 |
+
"loss": 0.407,
|
| 1076 |
+
"step": 65000
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 2.67,
|
| 1080 |
+
"learning_rate": 1.66568204041721e-06,
|
| 1081 |
+
"loss": 0.4128,
|
| 1082 |
+
"step": 65500
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 2.69,
|
| 1086 |
+
"learning_rate": 1.6402684973924382e-06,
|
| 1087 |
+
"loss": 0.4085,
|
| 1088 |
+
"step": 66000
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 2.69,
|
| 1092 |
+
"eval_accuracy": 0.795582329317269,
|
| 1093 |
+
"eval_loss": 0.5385735034942627,
|
| 1094 |
+
"eval_runtime": 3.1063,
|
| 1095 |
+
"eval_samples_per_second": 801.601,
|
| 1096 |
+
"eval_steps_per_second": 50.221,
|
| 1097 |
+
"step": 66000
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 2.71,
|
| 1101 |
+
"learning_rate": 1.6148040254237292e-06,
|
| 1102 |
+
"loss": 0.4036,
|
| 1103 |
+
"step": 66500
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 2.73,
|
| 1107 |
+
"learning_rate": 1.5893395534550195e-06,
|
| 1108 |
+
"loss": 0.4169,
|
| 1109 |
+
"step": 67000
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 2.75,
|
| 1113 |
+
"learning_rate": 1.5638750814863105e-06,
|
| 1114 |
+
"loss": 0.4301,
|
| 1115 |
+
"step": 67500
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 2.77,
|
| 1119 |
+
"learning_rate": 1.5384106095176013e-06,
|
| 1120 |
+
"loss": 0.4183,
|
| 1121 |
+
"step": 68000
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 2.77,
|
| 1125 |
+
"eval_accuracy": 0.8004016064257028,
|
| 1126 |
+
"eval_loss": 0.5275784730911255,
|
| 1127 |
+
"eval_runtime": 3.1078,
|
| 1128 |
+
"eval_samples_per_second": 801.199,
|
| 1129 |
+
"eval_steps_per_second": 50.196,
|
| 1130 |
+
"step": 68000
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 2.79,
|
| 1134 |
+
"learning_rate": 1.5129970664928293e-06,
|
| 1135 |
+
"loss": 0.4114,
|
| 1136 |
+
"step": 68500
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 2.81,
|
| 1140 |
+
"learning_rate": 1.48753259452412e-06,
|
| 1141 |
+
"loss": 0.4236,
|
| 1142 |
+
"step": 69000
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 2.83,
|
| 1146 |
+
"learning_rate": 1.4620681225554109e-06,
|
| 1147 |
+
"loss": 0.4179,
|
| 1148 |
+
"step": 69500
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 2.85,
|
| 1152 |
+
"learning_rate": 1.4366036505867014e-06,
|
| 1153 |
+
"loss": 0.4132,
|
| 1154 |
+
"step": 70000
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 2.85,
|
| 1158 |
+
"eval_accuracy": 0.7963855421686747,
|
| 1159 |
+
"eval_loss": 0.5376485586166382,
|
| 1160 |
+
"eval_runtime": 3.1089,
|
| 1161 |
+
"eval_samples_per_second": 800.921,
|
| 1162 |
+
"eval_steps_per_second": 50.178,
|
| 1163 |
+
"step": 70000
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 2.87,
|
| 1167 |
+
"learning_rate": 1.4111901075619296e-06,
|
| 1168 |
+
"loss": 0.4183,
|
| 1169 |
+
"step": 70500
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 2.89,
|
| 1173 |
+
"learning_rate": 1.3857765645371578e-06,
|
| 1174 |
+
"loss": 0.409,
|
| 1175 |
+
"step": 71000
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 2.91,
|
| 1179 |
+
"learning_rate": 1.3603120925684484e-06,
|
| 1180 |
+
"loss": 0.4113,
|
| 1181 |
+
"step": 71500
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 2.93,
|
| 1185 |
+
"learning_rate": 1.3348476205997394e-06,
|
| 1186 |
+
"loss": 0.4031,
|
| 1187 |
+
"step": 72000
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 2.93,
|
| 1191 |
+
"eval_accuracy": 0.7987951807228916,
|
| 1192 |
+
"eval_loss": 0.5440085530281067,
|
| 1193 |
+
"eval_runtime": 3.1049,
|
| 1194 |
+
"eval_samples_per_second": 801.949,
|
| 1195 |
+
"eval_steps_per_second": 50.243,
|
| 1196 |
+
"step": 72000
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 2.95,
|
| 1200 |
+
"learning_rate": 1.3093831486310302e-06,
|
| 1201 |
+
"loss": 0.4153,
|
| 1202 |
+
"step": 72500
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 2.97,
|
| 1206 |
+
"learning_rate": 1.2839186766623208e-06,
|
| 1207 |
+
"loss": 0.4092,
|
| 1208 |
+
"step": 73000
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 2.99,
|
| 1212 |
+
"learning_rate": 1.2584542046936116e-06,
|
| 1213 |
+
"loss": 0.4223,
|
| 1214 |
+
"step": 73500
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 3.01,
|
| 1218 |
+
"learning_rate": 1.2329897327249023e-06,
|
| 1219 |
+
"loss": 0.3561,
|
| 1220 |
+
"step": 74000
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 3.01,
|
| 1224 |
+
"eval_accuracy": 0.8032128514056225,
|
| 1225 |
+
"eval_loss": 0.5515400171279907,
|
| 1226 |
+
"eval_runtime": 3.1138,
|
| 1227 |
+
"eval_samples_per_second": 799.666,
|
| 1228 |
+
"eval_steps_per_second": 50.1,
|
| 1229 |
+
"step": 74000
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 3.04,
|
| 1233 |
+
"learning_rate": 1.2075252607561931e-06,
|
| 1234 |
+
"loss": 0.3444,
|
| 1235 |
+
"step": 74500
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 3.06,
|
| 1239 |
+
"learning_rate": 1.1821117177314211e-06,
|
| 1240 |
+
"loss": 0.3447,
|
| 1241 |
+
"step": 75000
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 3.08,
|
| 1245 |
+
"learning_rate": 1.156647245762712e-06,
|
| 1246 |
+
"loss": 0.3572,
|
| 1247 |
+
"step": 75500
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 3.1,
|
| 1251 |
+
"learning_rate": 1.1311827737940027e-06,
|
| 1252 |
+
"loss": 0.3453,
|
| 1253 |
+
"step": 76000
|
| 1254 |
+
},
|
| 1255 |
+
{
|
| 1256 |
+
"epoch": 3.1,
|
| 1257 |
+
"eval_accuracy": 0.8036144578313253,
|
| 1258 |
+
"eval_loss": 0.562953770160675,
|
| 1259 |
+
"eval_runtime": 3.1189,
|
| 1260 |
+
"eval_samples_per_second": 798.371,
|
| 1261 |
+
"eval_steps_per_second": 50.018,
|
| 1262 |
+
"step": 76000
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 3.12,
|
| 1266 |
+
"learning_rate": 1.1057183018252935e-06,
|
| 1267 |
+
"loss": 0.353,
|
| 1268 |
+
"step": 76500
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 3.14,
|
| 1272 |
+
"learning_rate": 1.0803047588005217e-06,
|
| 1273 |
+
"loss": 0.3489,
|
| 1274 |
+
"step": 77000
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 3.16,
|
| 1278 |
+
"learning_rate": 1.0548402868318123e-06,
|
| 1279 |
+
"loss": 0.3487,
|
| 1280 |
+
"step": 77500
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 3.18,
|
| 1284 |
+
"learning_rate": 1.029375814863103e-06,
|
| 1285 |
+
"loss": 0.3625,
|
| 1286 |
+
"step": 78000
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 3.18,
|
| 1290 |
+
"eval_accuracy": 0.8012048192771084,
|
| 1291 |
+
"eval_loss": 0.5594542026519775,
|
| 1292 |
+
"eval_runtime": 3.1009,
|
| 1293 |
+
"eval_samples_per_second": 802.99,
|
| 1294 |
+
"eval_steps_per_second": 50.308,
|
| 1295 |
+
"step": 78000
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 3.2,
|
| 1299 |
+
"learning_rate": 1.0039113428943938e-06,
|
| 1300 |
+
"loss": 0.3522,
|
| 1301 |
+
"step": 78500
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 3.22,
|
| 1305 |
+
"learning_rate": 9.784468709256846e-07,
|
| 1306 |
+
"loss": 0.3526,
|
| 1307 |
+
"step": 79000
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 3.24,
|
| 1311 |
+
"learning_rate": 9.530333279009127e-07,
|
| 1312 |
+
"loss": 0.3463,
|
| 1313 |
+
"step": 79500
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 3.26,
|
| 1317 |
+
"learning_rate": 9.275688559322035e-07,
|
| 1318 |
+
"loss": 0.3483,
|
| 1319 |
+
"step": 80000
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 3.26,
|
| 1323 |
+
"eval_accuracy": 0.8024096385542169,
|
| 1324 |
+
"eval_loss": 0.5727831125259399,
|
| 1325 |
+
"eval_runtime": 3.1106,
|
| 1326 |
+
"eval_samples_per_second": 800.501,
|
| 1327 |
+
"eval_steps_per_second": 50.152,
|
| 1328 |
+
"step": 80000
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 3.28,
|
| 1332 |
+
"learning_rate": 9.021043839634942e-07,
|
| 1333 |
+
"loss": 0.3614,
|
| 1334 |
+
"step": 80500
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 3.3,
|
| 1338 |
+
"learning_rate": 8.76639911994785e-07,
|
| 1339 |
+
"loss": 0.3464,
|
| 1340 |
+
"step": 81000
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 3.32,
|
| 1344 |
+
"learning_rate": 8.511754400260756e-07,
|
| 1345 |
+
"loss": 0.3534,
|
| 1346 |
+
"step": 81500
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 3.34,
|
| 1350 |
+
"learning_rate": 8.257618970013039e-07,
|
| 1351 |
+
"loss": 0.3452,
|
| 1352 |
+
"step": 82000
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 3.34,
|
| 1356 |
+
"eval_accuracy": 0.8032128514056225,
|
| 1357 |
+
"eval_loss": 0.5750517845153809,
|
| 1358 |
+
"eval_runtime": 3.0909,
|
| 1359 |
+
"eval_samples_per_second": 805.579,
|
| 1360 |
+
"eval_steps_per_second": 50.47,
|
| 1361 |
+
"step": 82000
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 3.36,
|
| 1365 |
+
"learning_rate": 8.002974250325945e-07,
|
| 1366 |
+
"loss": 0.3399,
|
| 1367 |
+
"step": 82500
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 3.38,
|
| 1371 |
+
"learning_rate": 7.748329530638854e-07,
|
| 1372 |
+
"loss": 0.3474,
|
| 1373 |
+
"step": 83000
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 3.4,
|
| 1377 |
+
"learning_rate": 7.493684810951761e-07,
|
| 1378 |
+
"loss": 0.3629,
|
| 1379 |
+
"step": 83500
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 3.42,
|
| 1383 |
+
"learning_rate": 7.239040091264668e-07,
|
| 1384 |
+
"loss": 0.3481,
|
| 1385 |
+
"step": 84000
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 3.42,
|
| 1389 |
+
"eval_accuracy": 0.7975903614457831,
|
| 1390 |
+
"eval_loss": 0.5842192769050598,
|
| 1391 |
+
"eval_runtime": 3.1136,
|
| 1392 |
+
"eval_samples_per_second": 799.707,
|
| 1393 |
+
"eval_steps_per_second": 50.102,
|
| 1394 |
+
"step": 84000
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 3.44,
|
| 1398 |
+
"learning_rate": 6.984395371577575e-07,
|
| 1399 |
+
"loss": 0.347,
|
| 1400 |
+
"step": 84500
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 3.46,
|
| 1404 |
+
"learning_rate": 6.729750651890484e-07,
|
| 1405 |
+
"loss": 0.3659,
|
| 1406 |
+
"step": 85000
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 3.48,
|
| 1410 |
+
"learning_rate": 6.47510593220339e-07,
|
| 1411 |
+
"loss": 0.3414,
|
| 1412 |
+
"step": 85500
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 3.5,
|
| 1416 |
+
"learning_rate": 6.220970501955672e-07,
|
| 1417 |
+
"loss": 0.3526,
|
| 1418 |
+
"step": 86000
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 3.5,
|
| 1422 |
+
"eval_accuracy": 0.7983935742971887,
|
| 1423 |
+
"eval_loss": 0.580718994140625,
|
| 1424 |
+
"eval_runtime": 3.0922,
|
| 1425 |
+
"eval_samples_per_second": 805.245,
|
| 1426 |
+
"eval_steps_per_second": 50.449,
|
| 1427 |
+
"step": 86000
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 3.52,
|
| 1431 |
+
"learning_rate": 5.966325782268579e-07,
|
| 1432 |
+
"loss": 0.3502,
|
| 1433 |
+
"step": 86500
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 3.54,
|
| 1437 |
+
"learning_rate": 5.712190352020861e-07,
|
| 1438 |
+
"loss": 0.3522,
|
| 1439 |
+
"step": 87000
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 3.57,
|
| 1443 |
+
"learning_rate": 5.457545632333768e-07,
|
| 1444 |
+
"loss": 0.3451,
|
| 1445 |
+
"step": 87500
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 3.59,
|
| 1449 |
+
"learning_rate": 5.202900912646676e-07,
|
| 1450 |
+
"loss": 0.3513,
|
| 1451 |
+
"step": 88000
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 3.59,
|
| 1455 |
+
"eval_accuracy": 0.8016064257028113,
|
| 1456 |
+
"eval_loss": 0.5780753493309021,
|
| 1457 |
+
"eval_runtime": 3.1077,
|
| 1458 |
+
"eval_samples_per_second": 801.223,
|
| 1459 |
+
"eval_steps_per_second": 50.197,
|
| 1460 |
+
"step": 88000
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 3.61,
|
| 1464 |
+
"learning_rate": 4.948256192959583e-07,
|
| 1465 |
+
"loss": 0.3386,
|
| 1466 |
+
"step": 88500
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 3.63,
|
| 1470 |
+
"learning_rate": 4.6936114732724906e-07,
|
| 1471 |
+
"loss": 0.3465,
|
| 1472 |
+
"step": 89000
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 3.65,
|
| 1476 |
+
"learning_rate": 4.4389667535853984e-07,
|
| 1477 |
+
"loss": 0.3474,
|
| 1478 |
+
"step": 89500
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 3.67,
|
| 1482 |
+
"learning_rate": 4.184322033898305e-07,
|
| 1483 |
+
"loss": 0.3288,
|
| 1484 |
+
"step": 90000
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 3.67,
|
| 1488 |
+
"eval_accuracy": 0.8008032128514057,
|
| 1489 |
+
"eval_loss": 0.5857793688774109,
|
| 1490 |
+
"eval_runtime": 3.11,
|
| 1491 |
+
"eval_samples_per_second": 800.647,
|
| 1492 |
+
"eval_steps_per_second": 50.161,
|
| 1493 |
+
"step": 90000
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 3.69,
|
| 1497 |
+
"learning_rate": 3.9306958930899614e-07,
|
| 1498 |
+
"loss": 0.3444,
|
| 1499 |
+
"step": 90500
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 3.71,
|
| 1503 |
+
"learning_rate": 3.676051173402868e-07,
|
| 1504 |
+
"loss": 0.3373,
|
| 1505 |
+
"step": 91000
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 3.73,
|
| 1509 |
+
"learning_rate": 3.421406453715776e-07,
|
| 1510 |
+
"loss": 0.3512,
|
| 1511 |
+
"step": 91500
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 3.75,
|
| 1515 |
+
"learning_rate": 3.166761734028683e-07,
|
| 1516 |
+
"loss": 0.3599,
|
| 1517 |
+
"step": 92000
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 3.75,
|
| 1521 |
+
"eval_accuracy": 0.8016064257028113,
|
| 1522 |
+
"eval_loss": 0.5658962726593018,
|
| 1523 |
+
"eval_runtime": 3.112,
|
| 1524 |
+
"eval_samples_per_second": 800.117,
|
| 1525 |
+
"eval_steps_per_second": 50.128,
|
| 1526 |
+
"step": 92000
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 3.77,
|
| 1530 |
+
"learning_rate": 2.9121170143415907e-07,
|
| 1531 |
+
"loss": 0.3535,
|
| 1532 |
+
"step": 92500
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 3.79,
|
| 1536 |
+
"learning_rate": 2.657472294654498e-07,
|
| 1537 |
+
"loss": 0.3404,
|
| 1538 |
+
"step": 93000
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 3.81,
|
| 1542 |
+
"learning_rate": 2.4028275749674054e-07,
|
| 1543 |
+
"loss": 0.3452,
|
| 1544 |
+
"step": 93500
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 3.83,
|
| 1548 |
+
"learning_rate": 2.148182855280313e-07,
|
| 1549 |
+
"loss": 0.3419,
|
| 1550 |
+
"step": 94000
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 3.83,
|
| 1554 |
+
"eval_accuracy": 0.8032128514056225,
|
| 1555 |
+
"eval_loss": 0.575123131275177,
|
| 1556 |
+
"eval_runtime": 3.1066,
|
| 1557 |
+
"eval_samples_per_second": 801.524,
|
| 1558 |
+
"eval_steps_per_second": 50.216,
|
| 1559 |
+
"step": 94000
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"epoch": 3.85,
|
| 1563 |
+
"learning_rate": 1.8935381355932206e-07,
|
| 1564 |
+
"loss": 0.3518,
|
| 1565 |
+
"step": 94500
|
| 1566 |
+
},
|
| 1567 |
+
{
|
| 1568 |
+
"epoch": 3.87,
|
| 1569 |
+
"learning_rate": 1.638893415906128e-07,
|
| 1570 |
+
"loss": 0.3444,
|
| 1571 |
+
"step": 95000
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 3.89,
|
| 1575 |
+
"learning_rate": 1.3842486962190353e-07,
|
| 1576 |
+
"loss": 0.3422,
|
| 1577 |
+
"step": 95500
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 3.91,
|
| 1581 |
+
"learning_rate": 1.1296039765319427e-07,
|
| 1582 |
+
"loss": 0.3502,
|
| 1583 |
+
"step": 96000
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"epoch": 3.91,
|
| 1587 |
+
"eval_accuracy": 0.8036144578313253,
|
| 1588 |
+
"eval_loss": 0.5685034394264221,
|
| 1589 |
+
"eval_runtime": 3.1416,
|
| 1590 |
+
"eval_samples_per_second": 792.595,
|
| 1591 |
+
"eval_steps_per_second": 49.657,
|
| 1592 |
+
"step": 96000
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 3.93,
|
| 1596 |
+
"learning_rate": 8.754685462842242e-08,
|
| 1597 |
+
"loss": 0.3373,
|
| 1598 |
+
"step": 96500
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 3.95,
|
| 1602 |
+
"learning_rate": 6.208238265971317e-08,
|
| 1603 |
+
"loss": 0.3587,
|
| 1604 |
+
"step": 97000
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 3.97,
|
| 1608 |
+
"learning_rate": 3.6617910691003915e-08,
|
| 1609 |
+
"loss": 0.3398,
|
| 1610 |
+
"step": 97500
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 3.99,
|
| 1614 |
+
"learning_rate": 1.1153438722294656e-08,
|
| 1615 |
+
"loss": 0.3443,
|
| 1616 |
+
"step": 98000
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 3.99,
|
| 1620 |
+
"eval_accuracy": 0.8016064257028113,
|
| 1621 |
+
"eval_loss": 0.5707261562347412,
|
| 1622 |
+
"eval_runtime": 3.1118,
|
| 1623 |
+
"eval_samples_per_second": 800.188,
|
| 1624 |
+
"eval_steps_per_second": 50.132,
|
| 1625 |
+
"step": 98000
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 4.0,
|
| 1629 |
+
"step": 98176,
|
| 1630 |
+
"total_flos": 1.1341991182001568e+16,
|
| 1631 |
+
"train_loss": 0.4772139054466879,
|
| 1632 |
+
"train_runtime": 9899.244,
|
| 1633 |
+
"train_samples_per_second": 158.68,
|
| 1634 |
+
"train_steps_per_second": 9.918
|
| 1635 |
+
}
|
| 1636 |
+
],
|
| 1637 |
+
"max_steps": 98176,
|
| 1638 |
+
"num_train_epochs": 4,
|
| 1639 |
+
"total_flos": 1.1341991182001568e+16,
|
| 1640 |
+
"trial_name": null,
|
| 1641 |
+
"trial_params": null
|
| 1642 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7c8c1af91ddb4f92dfce9caefc7452503cc47217750910d7a7ef5d0458152d1c
|
| 3 |
+
size 2863
|