Commit
•
d81d8f2
1
Parent(s):
a7a9a6d
adding model finetuned on XNLI
Browse files- all_results.json +14 -0
- config.json +39 -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 +427 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.7811244979919679,
|
| 4 |
+
"eval_loss": 0.5380504727363586,
|
| 5 |
+
"eval_runtime": 2.3737,
|
| 6 |
+
"eval_samples": 2490,
|
| 7 |
+
"eval_samples_per_second": 1048.988,
|
| 8 |
+
"eval_steps_per_second": 16.43,
|
| 9 |
+
"train_loss": 0.38933088829573315,
|
| 10 |
+
"train_runtime": 9866.0231,
|
| 11 |
+
"train_samples": 392702,
|
| 12 |
+
"train_samples_per_second": 159.214,
|
| 13 |
+
"train_steps_per_second": 2.488
|
| 14 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "dccuchile/bert-base-spanish-wwm-uncased",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"finetuning_task": "xnli",
|
| 9 |
+
"gradient_checkpointing": false,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"id2label": {
|
| 14 |
+
"0": "LABEL_0",
|
| 15 |
+
"1": "LABEL_1",
|
| 16 |
+
"2": "LABEL_2"
|
| 17 |
+
},
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"intermediate_size": 3072,
|
| 20 |
+
"label2id": {
|
| 21 |
+
"LABEL_0": 0,
|
| 22 |
+
"LABEL_1": 1,
|
| 23 |
+
"LABEL_2": 2
|
| 24 |
+
},
|
| 25 |
+
"layer_norm_eps": 1e-12,
|
| 26 |
+
"max_position_embeddings": 512,
|
| 27 |
+
"model_type": "bert",
|
| 28 |
+
"num_attention_heads": 12,
|
| 29 |
+
"num_hidden_layers": 12,
|
| 30 |
+
"output_past": true,
|
| 31 |
+
"pad_token_id": 1,
|
| 32 |
+
"position_embedding_type": "absolute",
|
| 33 |
+
"problem_type": "single_label_classification",
|
| 34 |
+
"torch_dtype": "float32",
|
| 35 |
+
"transformers_version": "4.12.5",
|
| 36 |
+
"type_vocab_size": 2,
|
| 37 |
+
"use_cache": true,
|
| 38 |
+
"vocab_size": 31002
|
| 39 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.7811244979919679,
|
| 4 |
+
"eval_loss": 0.5380504727363586,
|
| 5 |
+
"eval_runtime": 2.3737,
|
| 6 |
+
"eval_samples": 2490,
|
| 7 |
+
"eval_samples_per_second": 1048.988,
|
| 8 |
+
"eval_steps_per_second": 16.43
|
| 9 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:194383f123f451106990baab64f09ea3d13ca79c221c6b326019322d640ad9c7
|
| 3 |
+
size 439496877
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"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, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": false, "do_basic_tokenize": true, "never_split": null, "model_max_length": 512, "special_tokens_map_file": "/data/jcanete/cache/78141ed1e8dcc5ff370950397ca0d1c5c9da478f54ec14544187d8a93eff1a26.f982506b52498d4adb4bd491f593dc92b2ef6be61bfdbe9d30f53f963f9f5b66", "name_or_path": "dccuchile/bert-base-spanish-wwm-uncased", "tokenizer_class": "BertTokenizer"}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"train_loss": 0.38933088829573315,
|
| 4 |
+
"train_runtime": 9866.0231,
|
| 5 |
+
"train_samples": 392702,
|
| 6 |
+
"train_samples_per_second": 159.214,
|
| 7 |
+
"train_steps_per_second": 2.488
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,427 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.5380504727363586,
|
| 3 |
+
"best_model_checkpoint": "/data/jcanete/all_results/xnli/beto_uncased/epochs_4_bs_64_lr_5e-5/checkpoint-12000",
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"global_step": 24544,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.08,
|
| 12 |
+
"learning_rate": 4.898142112125163e-05,
|
| 13 |
+
"loss": 0.7866,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.16,
|
| 18 |
+
"learning_rate": 4.796487940026076e-05,
|
| 19 |
+
"loss": 0.703,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.24,
|
| 24 |
+
"learning_rate": 4.6946300521512387e-05,
|
| 25 |
+
"loss": 0.6809,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.33,
|
| 30 |
+
"learning_rate": 4.5927721642764015e-05,
|
| 31 |
+
"loss": 0.6555,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.33,
|
| 36 |
+
"eval_accuracy": 0.7497991967871486,
|
| 37 |
+
"eval_loss": 0.602264940738678,
|
| 38 |
+
"eval_runtime": 1.3367,
|
| 39 |
+
"eval_samples_per_second": 1862.863,
|
| 40 |
+
"eval_steps_per_second": 29.177,
|
| 41 |
+
"step": 2000
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.41,
|
| 45 |
+
"learning_rate": 4.490914276401565e-05,
|
| 46 |
+
"loss": 0.6351,
|
| 47 |
+
"step": 2500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.49,
|
| 51 |
+
"learning_rate": 4.389056388526728e-05,
|
| 52 |
+
"loss": 0.6267,
|
| 53 |
+
"step": 3000
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.57,
|
| 57 |
+
"learning_rate": 4.2871985006518906e-05,
|
| 58 |
+
"loss": 0.62,
|
| 59 |
+
"step": 3500
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.65,
|
| 63 |
+
"learning_rate": 4.185340612777054e-05,
|
| 64 |
+
"loss": 0.6094,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.65,
|
| 69 |
+
"eval_accuracy": 0.7678714859437751,
|
| 70 |
+
"eval_loss": 0.5793344378471375,
|
| 71 |
+
"eval_runtime": 2.4971,
|
| 72 |
+
"eval_samples_per_second": 997.156,
|
| 73 |
+
"eval_steps_per_second": 15.618,
|
| 74 |
+
"step": 4000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.73,
|
| 78 |
+
"learning_rate": 4.083482724902217e-05,
|
| 79 |
+
"loss": 0.5983,
|
| 80 |
+
"step": 4500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.81,
|
| 84 |
+
"learning_rate": 3.9816248370273797e-05,
|
| 85 |
+
"loss": 0.5886,
|
| 86 |
+
"step": 5000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.9,
|
| 90 |
+
"learning_rate": 3.880174380704042e-05,
|
| 91 |
+
"loss": 0.5871,
|
| 92 |
+
"step": 5500
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.98,
|
| 96 |
+
"learning_rate": 3.7783164928292046e-05,
|
| 97 |
+
"loss": 0.5771,
|
| 98 |
+
"step": 6000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.98,
|
| 102 |
+
"eval_accuracy": 0.7811244979919679,
|
| 103 |
+
"eval_loss": 0.5455663800239563,
|
| 104 |
+
"eval_runtime": 2.4323,
|
| 105 |
+
"eval_samples_per_second": 1023.703,
|
| 106 |
+
"eval_steps_per_second": 16.034,
|
| 107 |
+
"step": 6000
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.06,
|
| 111 |
+
"learning_rate": 3.676458604954368e-05,
|
| 112 |
+
"loss": 0.4853,
|
| 113 |
+
"step": 6500
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.14,
|
| 117 |
+
"learning_rate": 3.574600717079531e-05,
|
| 118 |
+
"loss": 0.4519,
|
| 119 |
+
"step": 7000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.22,
|
| 123 |
+
"learning_rate": 3.472742829204694e-05,
|
| 124 |
+
"loss": 0.4597,
|
| 125 |
+
"step": 7500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 1.3,
|
| 129 |
+
"learning_rate": 3.3708849413298565e-05,
|
| 130 |
+
"loss": 0.4484,
|
| 131 |
+
"step": 8000
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 1.3,
|
| 135 |
+
"eval_accuracy": 0.7899598393574297,
|
| 136 |
+
"eval_loss": 0.5768685340881348,
|
| 137 |
+
"eval_runtime": 1.3402,
|
| 138 |
+
"eval_samples_per_second": 1857.904,
|
| 139 |
+
"eval_steps_per_second": 29.1,
|
| 140 |
+
"step": 8000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.39,
|
| 144 |
+
"learning_rate": 3.26902705345502e-05,
|
| 145 |
+
"loss": 0.4567,
|
| 146 |
+
"step": 8500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 1.47,
|
| 150 |
+
"learning_rate": 3.167372881355932e-05,
|
| 151 |
+
"loss": 0.4483,
|
| 152 |
+
"step": 9000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 1.55,
|
| 156 |
+
"learning_rate": 3.065514993481095e-05,
|
| 157 |
+
"loss": 0.446,
|
| 158 |
+
"step": 9500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.63,
|
| 162 |
+
"learning_rate": 2.9636571056062584e-05,
|
| 163 |
+
"loss": 0.4513,
|
| 164 |
+
"step": 10000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.63,
|
| 168 |
+
"eval_accuracy": 0.7738955823293173,
|
| 169 |
+
"eval_loss": 0.5736198425292969,
|
| 170 |
+
"eval_runtime": 1.3279,
|
| 171 |
+
"eval_samples_per_second": 1875.149,
|
| 172 |
+
"eval_steps_per_second": 29.37,
|
| 173 |
+
"step": 10000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 1.71,
|
| 177 |
+
"learning_rate": 2.8617992177314212e-05,
|
| 178 |
+
"loss": 0.4511,
|
| 179 |
+
"step": 10500
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 1.79,
|
| 183 |
+
"learning_rate": 2.759941329856584e-05,
|
| 184 |
+
"loss": 0.4423,
|
| 185 |
+
"step": 11000
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.87,
|
| 189 |
+
"learning_rate": 2.6580834419817475e-05,
|
| 190 |
+
"loss": 0.4472,
|
| 191 |
+
"step": 11500
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 1.96,
|
| 195 |
+
"learning_rate": 2.5564292698826596e-05,
|
| 196 |
+
"loss": 0.4362,
|
| 197 |
+
"step": 12000
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 1.96,
|
| 201 |
+
"eval_accuracy": 0.7811244979919679,
|
| 202 |
+
"eval_loss": 0.5380504727363586,
|
| 203 |
+
"eval_runtime": 1.33,
|
| 204 |
+
"eval_samples_per_second": 1872.162,
|
| 205 |
+
"eval_steps_per_second": 29.323,
|
| 206 |
+
"step": 12000
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 2.04,
|
| 210 |
+
"learning_rate": 2.4545713820078228e-05,
|
| 211 |
+
"loss": 0.3786,
|
| 212 |
+
"step": 12500
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 2.12,
|
| 216 |
+
"learning_rate": 2.3527134941329856e-05,
|
| 217 |
+
"loss": 0.2879,
|
| 218 |
+
"step": 13000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 2.2,
|
| 222 |
+
"learning_rate": 2.2508556062581487e-05,
|
| 223 |
+
"loss": 0.2885,
|
| 224 |
+
"step": 13500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.28,
|
| 228 |
+
"learning_rate": 2.148997718383312e-05,
|
| 229 |
+
"loss": 0.2909,
|
| 230 |
+
"step": 14000
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 2.28,
|
| 234 |
+
"eval_accuracy": 0.7835341365461848,
|
| 235 |
+
"eval_loss": 0.6650500297546387,
|
| 236 |
+
"eval_runtime": 2.4288,
|
| 237 |
+
"eval_samples_per_second": 1025.198,
|
| 238 |
+
"eval_steps_per_second": 16.057,
|
| 239 |
+
"step": 14000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 2.36,
|
| 243 |
+
"learning_rate": 2.0473435462842243e-05,
|
| 244 |
+
"loss": 0.2991,
|
| 245 |
+
"step": 14500
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 2.44,
|
| 249 |
+
"learning_rate": 1.945485658409387e-05,
|
| 250 |
+
"loss": 0.2865,
|
| 251 |
+
"step": 15000
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 2.53,
|
| 255 |
+
"learning_rate": 1.8436277705345503e-05,
|
| 256 |
+
"loss": 0.2947,
|
| 257 |
+
"step": 15500
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 2.61,
|
| 261 |
+
"learning_rate": 1.741769882659713e-05,
|
| 262 |
+
"loss": 0.3002,
|
| 263 |
+
"step": 16000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 2.61,
|
| 267 |
+
"eval_accuracy": 0.7807228915662651,
|
| 268 |
+
"eval_loss": 0.6428889632225037,
|
| 269 |
+
"eval_runtime": 2.4279,
|
| 270 |
+
"eval_samples_per_second": 1025.584,
|
| 271 |
+
"eval_steps_per_second": 16.063,
|
| 272 |
+
"step": 16000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 2.69,
|
| 276 |
+
"learning_rate": 1.6399119947848762e-05,
|
| 277 |
+
"loss": 0.2916,
|
| 278 |
+
"step": 16500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 2.77,
|
| 282 |
+
"learning_rate": 1.5380541069100394e-05,
|
| 283 |
+
"loss": 0.2987,
|
| 284 |
+
"step": 17000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 2.85,
|
| 288 |
+
"learning_rate": 1.4361962190352022e-05,
|
| 289 |
+
"loss": 0.3005,
|
| 290 |
+
"step": 17500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 2.93,
|
| 294 |
+
"learning_rate": 1.3343383311603652e-05,
|
| 295 |
+
"loss": 0.2916,
|
| 296 |
+
"step": 18000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 2.93,
|
| 300 |
+
"eval_accuracy": 0.7779116465863454,
|
| 301 |
+
"eval_loss": 0.6353530883789062,
|
| 302 |
+
"eval_runtime": 1.3158,
|
| 303 |
+
"eval_samples_per_second": 1892.365,
|
| 304 |
+
"eval_steps_per_second": 29.639,
|
| 305 |
+
"step": 18000
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 3.01,
|
| 309 |
+
"learning_rate": 1.2328878748370275e-05,
|
| 310 |
+
"loss": 0.2756,
|
| 311 |
+
"step": 18500
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 3.1,
|
| 315 |
+
"learning_rate": 1.1310299869621905e-05,
|
| 316 |
+
"loss": 0.1773,
|
| 317 |
+
"step": 19000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 3.18,
|
| 321 |
+
"learning_rate": 1.0291720990873534e-05,
|
| 322 |
+
"loss": 0.1788,
|
| 323 |
+
"step": 19500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 3.26,
|
| 327 |
+
"learning_rate": 9.273142112125164e-06,
|
| 328 |
+
"loss": 0.1801,
|
| 329 |
+
"step": 20000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 3.26,
|
| 333 |
+
"eval_accuracy": 0.7811244979919679,
|
| 334 |
+
"eval_loss": 0.7538436055183411,
|
| 335 |
+
"eval_runtime": 1.3281,
|
| 336 |
+
"eval_samples_per_second": 1874.884,
|
| 337 |
+
"eval_steps_per_second": 29.366,
|
| 338 |
+
"step": 20000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 3.34,
|
| 342 |
+
"learning_rate": 8.254563233376792e-06,
|
| 343 |
+
"loss": 0.176,
|
| 344 |
+
"step": 20500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.42,
|
| 348 |
+
"learning_rate": 7.23802151238592e-06,
|
| 349 |
+
"loss": 0.1801,
|
| 350 |
+
"step": 21000
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 3.5,
|
| 354 |
+
"learning_rate": 6.219442633637549e-06,
|
| 355 |
+
"loss": 0.1761,
|
| 356 |
+
"step": 21500
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 3.59,
|
| 360 |
+
"learning_rate": 5.200863754889179e-06,
|
| 361 |
+
"loss": 0.1795,
|
| 362 |
+
"step": 22000
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 3.59,
|
| 366 |
+
"eval_accuracy": 0.7891566265060241,
|
| 367 |
+
"eval_loss": 0.7428993582725525,
|
| 368 |
+
"eval_runtime": 2.4875,
|
| 369 |
+
"eval_samples_per_second": 1000.986,
|
| 370 |
+
"eval_steps_per_second": 15.678,
|
| 371 |
+
"step": 22000
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 3.67,
|
| 375 |
+
"learning_rate": 4.182284876140809e-06,
|
| 376 |
+
"loss": 0.1734,
|
| 377 |
+
"step": 22500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 3.75,
|
| 381 |
+
"learning_rate": 3.1637059973924383e-06,
|
| 382 |
+
"loss": 0.1759,
|
| 383 |
+
"step": 23000
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 3.83,
|
| 387 |
+
"learning_rate": 2.1471642764015647e-06,
|
| 388 |
+
"loss": 0.1789,
|
| 389 |
+
"step": 23500
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 3.91,
|
| 393 |
+
"learning_rate": 1.1285853976531943e-06,
|
| 394 |
+
"loss": 0.1723,
|
| 395 |
+
"step": 24000
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 3.91,
|
| 399 |
+
"eval_accuracy": 0.7895582329317269,
|
| 400 |
+
"eval_loss": 0.7440813183784485,
|
| 401 |
+
"eval_runtime": 2.4411,
|
| 402 |
+
"eval_samples_per_second": 1020.029,
|
| 403 |
+
"eval_steps_per_second": 15.976,
|
| 404 |
+
"step": 24000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 3.99,
|
| 408 |
+
"learning_rate": 1.1000651890482399e-07,
|
| 409 |
+
"loss": 0.17,
|
| 410 |
+
"step": 24500
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 4.0,
|
| 414 |
+
"step": 24544,
|
| 415 |
+
"total_flos": 8.969723878657651e+16,
|
| 416 |
+
"train_loss": 0.38933088829573315,
|
| 417 |
+
"train_runtime": 9866.0231,
|
| 418 |
+
"train_samples_per_second": 159.214,
|
| 419 |
+
"train_steps_per_second": 2.488
|
| 420 |
+
}
|
| 421 |
+
],
|
| 422 |
+
"max_steps": 24544,
|
| 423 |
+
"num_train_epochs": 4,
|
| 424 |
+
"total_flos": 8.969723878657651e+16,
|
| 425 |
+
"trial_name": null,
|
| 426 |
+
"trial_params": null
|
| 427 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:47684929a44c62c2e844237c0bade0398264177902f74897906161d557c68c6c
|
| 3 |
+
size 2863
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|