Upload 13 files
Browse files- config.json +53 -0
- merges.txt +0 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +15 -0
- tokenizer.json +0 -0
- tokenizer_config.json +15 -0
- trainer_state.json +478 -0
- training_args.bin +3 -0
- vocab.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,53 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/apdcephfs/share_916081/effidit_shared_data/yafuli/Codes/LLM-results/classfication/models/longformer_base_4096",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LongformerForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_mode": "longformer",
|
| 7 |
+
"attention_probs_dropout_prob": 0.1,
|
| 8 |
+
"attention_window": [
|
| 9 |
+
512,
|
| 10 |
+
512,
|
| 11 |
+
512,
|
| 12 |
+
512,
|
| 13 |
+
512,
|
| 14 |
+
512,
|
| 15 |
+
512,
|
| 16 |
+
512,
|
| 17 |
+
512,
|
| 18 |
+
512,
|
| 19 |
+
512,
|
| 20 |
+
512
|
| 21 |
+
],
|
| 22 |
+
"bos_token_id": 0,
|
| 23 |
+
"eos_token_id": 2,
|
| 24 |
+
"gradient_checkpointing": false,
|
| 25 |
+
"hidden_act": "gelu",
|
| 26 |
+
"hidden_dropout_prob": 0.1,
|
| 27 |
+
"hidden_size": 768,
|
| 28 |
+
"id2label": {
|
| 29 |
+
"0": 0,
|
| 30 |
+
"1": 1
|
| 31 |
+
},
|
| 32 |
+
"ignore_attention_mask": false,
|
| 33 |
+
"initializer_range": 0.02,
|
| 34 |
+
"intermediate_size": 3072,
|
| 35 |
+
"label2id": {
|
| 36 |
+
"0": 0,
|
| 37 |
+
"1": 1
|
| 38 |
+
},
|
| 39 |
+
"layer_norm_eps": 1e-05,
|
| 40 |
+
"max_position_embeddings": 4098,
|
| 41 |
+
"model_type": "longformer",
|
| 42 |
+
"num_attention_heads": 12,
|
| 43 |
+
"num_hidden_layers": 12,
|
| 44 |
+
"onnx_export": false,
|
| 45 |
+
"pad_token_id": 1,
|
| 46 |
+
"position_embedding_type": "absolute",
|
| 47 |
+
"problem_type": "single_label_classification",
|
| 48 |
+
"sep_token_id": 2,
|
| 49 |
+
"torch_dtype": "float32",
|
| 50 |
+
"transformers_version": "4.27.4",
|
| 51 |
+
"type_vocab_size": 1,
|
| 52 |
+
"vocab_size": 50265
|
| 53 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f6b9b31fc7deb68db3738f6a96fe2b0adc075b9e0592418ce49056340664b658
|
| 3 |
+
size 1189446589
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9a8853ee3563ba5d6f251dba1421a028dd2251ff80c0ba110650dfed5a53766e
|
| 3 |
+
size 594737055
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bae425a9a22152b4243f1294e3c4f6f1ac182f90d2c74645f2626619079787a7
|
| 3 |
+
size 21579
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4eb96d3c159ec49bfc76551332a7f52392c99a69b649688e4dc3f1979f6d527b
|
| 3 |
+
size 559
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5d95c1e1c6ec2c56daac9bd96a1fee70862322dcd845123edfab43941f796769
|
| 3 |
+
size 623
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"cls_token": "<s>",
|
| 4 |
+
"eos_token": "</s>",
|
| 5 |
+
"mask_token": {
|
| 6 |
+
"content": "<mask>",
|
| 7 |
+
"lstrip": true,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "</s>",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"bos_token": "<s>",
|
| 4 |
+
"cls_token": "<s>",
|
| 5 |
+
"eos_token": "</s>",
|
| 6 |
+
"errors": "replace",
|
| 7 |
+
"mask_token": "<mask>",
|
| 8 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 9 |
+
"pad_token": "<pad>",
|
| 10 |
+
"sep_token": "</s>",
|
| 11 |
+
"special_tokens_map_file": null,
|
| 12 |
+
"tokenizer_class": "LongformerTokenizer",
|
| 13 |
+
"trim_offsets": true,
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,478 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.262811541557312,
|
| 3 |
+
"best_model_checkpoint": "./output_samples_202302241350_lfbase/checkpoint-6000",
|
| 4 |
+
"epoch": 4.245465071401004,
|
| 5 |
+
"global_step": 22000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.1,
|
| 12 |
+
"learning_rate": 2.94245465071401e-05,
|
| 13 |
+
"loss": 0.2869,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.19,
|
| 18 |
+
"learning_rate": 2.8845619451949058e-05,
|
| 19 |
+
"loss": 0.1797,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.19,
|
| 24 |
+
"eval_accuracy": 0.8138957619667053,
|
| 25 |
+
"eval_loss": 0.49187058210372925,
|
| 26 |
+
"eval_runtime": 490.7956,
|
| 27 |
+
"eval_samples_per_second": 119.795,
|
| 28 |
+
"eval_steps_per_second": 1.872,
|
| 29 |
+
"step": 1000
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.29,
|
| 33 |
+
"learning_rate": 2.826669239675801e-05,
|
| 34 |
+
"loss": 0.1589,
|
| 35 |
+
"step": 1500
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 0.39,
|
| 39 |
+
"learning_rate": 2.7687765341566965e-05,
|
| 40 |
+
"loss": 0.1369,
|
| 41 |
+
"step": 2000
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.39,
|
| 45 |
+
"eval_accuracy": 0.8265668749809265,
|
| 46 |
+
"eval_loss": 0.4808556139469147,
|
| 47 |
+
"eval_runtime": 472.7272,
|
| 48 |
+
"eval_samples_per_second": 124.374,
|
| 49 |
+
"eval_steps_per_second": 1.944,
|
| 50 |
+
"step": 2000
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.48,
|
| 54 |
+
"learning_rate": 2.71099961404863e-05,
|
| 55 |
+
"loss": 0.128,
|
| 56 |
+
"step": 2500
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.58,
|
| 60 |
+
"learning_rate": 2.6531069085295255e-05,
|
| 61 |
+
"loss": 0.1274,
|
| 62 |
+
"step": 3000
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.58,
|
| 66 |
+
"eval_accuracy": 0.8054426312446594,
|
| 67 |
+
"eval_loss": 0.6082525253295898,
|
| 68 |
+
"eval_runtime": 485.5844,
|
| 69 |
+
"eval_samples_per_second": 121.081,
|
| 70 |
+
"eval_steps_per_second": 1.893,
|
| 71 |
+
"step": 3000
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 0.68,
|
| 75 |
+
"learning_rate": 2.5953299884214592e-05,
|
| 76 |
+
"loss": 0.1161,
|
| 77 |
+
"step": 3500
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.77,
|
| 81 |
+
"learning_rate": 2.5374372829023545e-05,
|
| 82 |
+
"loss": 0.1109,
|
| 83 |
+
"step": 4000
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.77,
|
| 87 |
+
"eval_accuracy": 0.8235734105110168,
|
| 88 |
+
"eval_loss": 0.510552704334259,
|
| 89 |
+
"eval_runtime": 488.158,
|
| 90 |
+
"eval_samples_per_second": 120.443,
|
| 91 |
+
"eval_steps_per_second": 1.883,
|
| 92 |
+
"step": 4000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.87,
|
| 96 |
+
"learning_rate": 2.47954457738325e-05,
|
| 97 |
+
"loss": 0.1078,
|
| 98 |
+
"step": 4500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.96,
|
| 102 |
+
"learning_rate": 2.4216518718641452e-05,
|
| 103 |
+
"loss": 0.1036,
|
| 104 |
+
"step": 5000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.96,
|
| 108 |
+
"eval_accuracy": 0.8834084272384644,
|
| 109 |
+
"eval_loss": 0.312045156955719,
|
| 110 |
+
"eval_runtime": 486.8674,
|
| 111 |
+
"eval_samples_per_second": 120.762,
|
| 112 |
+
"eval_steps_per_second": 1.888,
|
| 113 |
+
"step": 5000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.06,
|
| 117 |
+
"learning_rate": 2.3637591663450405e-05,
|
| 118 |
+
"loss": 0.0823,
|
| 119 |
+
"step": 5500
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.16,
|
| 123 |
+
"learning_rate": 2.305866460825936e-05,
|
| 124 |
+
"loss": 0.0718,
|
| 125 |
+
"step": 6000
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 1.16,
|
| 129 |
+
"eval_accuracy": 0.9076281785964966,
|
| 130 |
+
"eval_loss": 0.262811541557312,
|
| 131 |
+
"eval_runtime": 488.7776,
|
| 132 |
+
"eval_samples_per_second": 120.29,
|
| 133 |
+
"eval_steps_per_second": 1.88,
|
| 134 |
+
"step": 6000
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.25,
|
| 138 |
+
"learning_rate": 2.2479737553068312e-05,
|
| 139 |
+
"loss": 0.0714,
|
| 140 |
+
"step": 6500
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.35,
|
| 144 |
+
"learning_rate": 2.190081049787727e-05,
|
| 145 |
+
"loss": 0.0729,
|
| 146 |
+
"step": 7000
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 1.35,
|
| 150 |
+
"eval_accuracy": 0.8230121731758118,
|
| 151 |
+
"eval_loss": 0.6299644112586975,
|
| 152 |
+
"eval_runtime": 480.4556,
|
| 153 |
+
"eval_samples_per_second": 122.373,
|
| 154 |
+
"eval_steps_per_second": 1.913,
|
| 155 |
+
"step": 7000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 1.45,
|
| 159 |
+
"learning_rate": 2.1321883442686222e-05,
|
| 160 |
+
"loss": 0.0687,
|
| 161 |
+
"step": 7500
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 1.54,
|
| 165 |
+
"learning_rate": 2.0742956387495175e-05,
|
| 166 |
+
"loss": 0.0664,
|
| 167 |
+
"step": 8000
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 1.54,
|
| 171 |
+
"eval_accuracy": 0.8701080083847046,
|
| 172 |
+
"eval_loss": 0.5767059922218323,
|
| 173 |
+
"eval_runtime": 472.3986,
|
| 174 |
+
"eval_samples_per_second": 124.461,
|
| 175 |
+
"eval_steps_per_second": 1.945,
|
| 176 |
+
"step": 8000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.64,
|
| 180 |
+
"learning_rate": 2.0164029332304132e-05,
|
| 181 |
+
"loss": 0.0665,
|
| 182 |
+
"step": 8500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 1.74,
|
| 186 |
+
"learning_rate": 1.9585102277113085e-05,
|
| 187 |
+
"loss": 0.0622,
|
| 188 |
+
"step": 9000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 1.74,
|
| 192 |
+
"eval_accuracy": 0.8938174843788147,
|
| 193 |
+
"eval_loss": 0.46610942482948303,
|
| 194 |
+
"eval_runtime": 472.4306,
|
| 195 |
+
"eval_samples_per_second": 124.452,
|
| 196 |
+
"eval_steps_per_second": 1.945,
|
| 197 |
+
"step": 9000
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 1.83,
|
| 201 |
+
"learning_rate": 1.9007333076032422e-05,
|
| 202 |
+
"loss": 0.0639,
|
| 203 |
+
"step": 9500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 1.93,
|
| 207 |
+
"learning_rate": 1.8428406020841375e-05,
|
| 208 |
+
"loss": 0.0638,
|
| 209 |
+
"step": 10000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 1.93,
|
| 213 |
+
"eval_accuracy": 0.913428008556366,
|
| 214 |
+
"eval_loss": 0.27155911922454834,
|
| 215 |
+
"eval_runtime": 484.5228,
|
| 216 |
+
"eval_samples_per_second": 121.346,
|
| 217 |
+
"eval_steps_per_second": 1.897,
|
| 218 |
+
"step": 10000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 2.03,
|
| 222 |
+
"learning_rate": 1.784947896565033e-05,
|
| 223 |
+
"loss": 0.0564,
|
| 224 |
+
"step": 10500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.12,
|
| 228 |
+
"learning_rate": 1.7270551910459282e-05,
|
| 229 |
+
"loss": 0.0385,
|
| 230 |
+
"step": 11000
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 2.12,
|
| 234 |
+
"eval_accuracy": 0.9116081595420837,
|
| 235 |
+
"eval_loss": 0.38816747069358826,
|
| 236 |
+
"eval_runtime": 488.2826,
|
| 237 |
+
"eval_samples_per_second": 120.412,
|
| 238 |
+
"eval_steps_per_second": 1.882,
|
| 239 |
+
"step": 11000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 2.22,
|
| 243 |
+
"learning_rate": 1.669162485526824e-05,
|
| 244 |
+
"loss": 0.0412,
|
| 245 |
+
"step": 11500
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 2.32,
|
| 249 |
+
"learning_rate": 1.6113855654187572e-05,
|
| 250 |
+
"loss": 0.0406,
|
| 251 |
+
"step": 12000
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 2.32,
|
| 255 |
+
"eval_accuracy": 0.863440752029419,
|
| 256 |
+
"eval_loss": 0.6258434653282166,
|
| 257 |
+
"eval_runtime": 473.3305,
|
| 258 |
+
"eval_samples_per_second": 124.216,
|
| 259 |
+
"eval_steps_per_second": 1.942,
|
| 260 |
+
"step": 12000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 2.41,
|
| 264 |
+
"learning_rate": 1.553492859899653e-05,
|
| 265 |
+
"loss": 0.0391,
|
| 266 |
+
"step": 12500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 2.51,
|
| 270 |
+
"learning_rate": 1.495600154380548e-05,
|
| 271 |
+
"loss": 0.037,
|
| 272 |
+
"step": 13000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 2.51,
|
| 276 |
+
"eval_accuracy": 0.891011118888855,
|
| 277 |
+
"eval_loss": 0.48729509115219116,
|
| 278 |
+
"eval_runtime": 473.8644,
|
| 279 |
+
"eval_samples_per_second": 124.076,
|
| 280 |
+
"eval_steps_per_second": 1.939,
|
| 281 |
+
"step": 13000
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 2.61,
|
| 285 |
+
"learning_rate": 1.4377074488614436e-05,
|
| 286 |
+
"loss": 0.04,
|
| 287 |
+
"step": 13500
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 2.7,
|
| 291 |
+
"learning_rate": 1.3798147433423389e-05,
|
| 292 |
+
"loss": 0.0382,
|
| 293 |
+
"step": 14000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 2.7,
|
| 297 |
+
"eval_accuracy": 0.8833404183387756,
|
| 298 |
+
"eval_loss": 0.5917666554450989,
|
| 299 |
+
"eval_runtime": 473.8023,
|
| 300 |
+
"eval_samples_per_second": 124.092,
|
| 301 |
+
"eval_steps_per_second": 1.94,
|
| 302 |
+
"step": 14000
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 2.8,
|
| 306 |
+
"learning_rate": 1.3220378232342726e-05,
|
| 307 |
+
"loss": 0.0392,
|
| 308 |
+
"step": 14500
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 2.89,
|
| 312 |
+
"learning_rate": 1.2642609031262061e-05,
|
| 313 |
+
"loss": 0.0367,
|
| 314 |
+
"step": 15000
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 2.89,
|
| 318 |
+
"eval_accuracy": 0.8793604969978333,
|
| 319 |
+
"eval_loss": 0.569683313369751,
|
| 320 |
+
"eval_runtime": 474.4494,
|
| 321 |
+
"eval_samples_per_second": 123.923,
|
| 322 |
+
"eval_steps_per_second": 1.937,
|
| 323 |
+
"step": 15000
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 2.99,
|
| 327 |
+
"learning_rate": 1.2063681976071016e-05,
|
| 328 |
+
"loss": 0.0358,
|
| 329 |
+
"step": 15500
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 3.09,
|
| 333 |
+
"learning_rate": 1.148475492087997e-05,
|
| 334 |
+
"loss": 0.0216,
|
| 335 |
+
"step": 16000
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 3.09,
|
| 339 |
+
"eval_accuracy": 0.8624373078346252,
|
| 340 |
+
"eval_loss": 0.8281469345092773,
|
| 341 |
+
"eval_runtime": 473.8459,
|
| 342 |
+
"eval_samples_per_second": 124.08,
|
| 343 |
+
"eval_steps_per_second": 1.939,
|
| 344 |
+
"step": 16000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.18,
|
| 348 |
+
"learning_rate": 1.0905827865688923e-05,
|
| 349 |
+
"loss": 0.0211,
|
| 350 |
+
"step": 16500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 3.28,
|
| 354 |
+
"learning_rate": 1.032805866460826e-05,
|
| 355 |
+
"loss": 0.0227,
|
| 356 |
+
"step": 17000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 3.28,
|
| 360 |
+
"eval_accuracy": 0.8539331555366516,
|
| 361 |
+
"eval_loss": 0.9032775163650513,
|
| 362 |
+
"eval_runtime": 474.5611,
|
| 363 |
+
"eval_samples_per_second": 123.893,
|
| 364 |
+
"eval_steps_per_second": 1.937,
|
| 365 |
+
"step": 17000
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 3.38,
|
| 369 |
+
"learning_rate": 9.749131609417213e-06,
|
| 370 |
+
"loss": 0.0225,
|
| 371 |
+
"step": 17500
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 3.47,
|
| 375 |
+
"learning_rate": 9.170204554226166e-06,
|
| 376 |
+
"loss": 0.0216,
|
| 377 |
+
"step": 18000
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 3.47,
|
| 381 |
+
"eval_accuracy": 0.888221800327301,
|
| 382 |
+
"eval_loss": 0.6849754452705383,
|
| 383 |
+
"eval_runtime": 487.296,
|
| 384 |
+
"eval_samples_per_second": 120.656,
|
| 385 |
+
"eval_steps_per_second": 1.886,
|
| 386 |
+
"step": 18000
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 3.57,
|
| 390 |
+
"learning_rate": 8.591277499035123e-06,
|
| 391 |
+
"loss": 0.0217,
|
| 392 |
+
"step": 18500
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 3.67,
|
| 396 |
+
"learning_rate": 8.012350443844076e-06,
|
| 397 |
+
"loss": 0.0199,
|
| 398 |
+
"step": 19000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 3.67,
|
| 402 |
+
"eval_accuracy": 0.8972702026367188,
|
| 403 |
+
"eval_loss": 0.6052428483963013,
|
| 404 |
+
"eval_runtime": 490.4243,
|
| 405 |
+
"eval_samples_per_second": 119.886,
|
| 406 |
+
"eval_steps_per_second": 1.874,
|
| 407 |
+
"step": 19000
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 3.76,
|
| 411 |
+
"learning_rate": 7.43342338865303e-06,
|
| 412 |
+
"loss": 0.0207,
|
| 413 |
+
"step": 19500
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 3.86,
|
| 417 |
+
"learning_rate": 6.855654187572366e-06,
|
| 418 |
+
"loss": 0.0206,
|
| 419 |
+
"step": 20000
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 3.86,
|
| 423 |
+
"eval_accuracy": 0.8873884081840515,
|
| 424 |
+
"eval_loss": 0.6865639090538025,
|
| 425 |
+
"eval_runtime": 487.3484,
|
| 426 |
+
"eval_samples_per_second": 120.643,
|
| 427 |
+
"eval_steps_per_second": 1.886,
|
| 428 |
+
"step": 20000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 3.96,
|
| 432 |
+
"learning_rate": 6.276727132381321e-06,
|
| 433 |
+
"loss": 0.0187,
|
| 434 |
+
"step": 20500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 4.05,
|
| 438 |
+
"learning_rate": 5.697800077190274e-06,
|
| 439 |
+
"loss": 0.0148,
|
| 440 |
+
"step": 21000
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 4.05,
|
| 444 |
+
"eval_accuracy": 0.8811633586883545,
|
| 445 |
+
"eval_loss": 0.9089483618736267,
|
| 446 |
+
"eval_runtime": 487.5135,
|
| 447 |
+
"eval_samples_per_second": 120.602,
|
| 448 |
+
"eval_steps_per_second": 1.885,
|
| 449 |
+
"step": 21000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 4.15,
|
| 453 |
+
"learning_rate": 5.118873021999228e-06,
|
| 454 |
+
"loss": 0.0107,
|
| 455 |
+
"step": 21500
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 4.25,
|
| 459 |
+
"learning_rate": 4.5399459668081825e-06,
|
| 460 |
+
"loss": 0.0121,
|
| 461 |
+
"step": 22000
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 4.25,
|
| 465 |
+
"eval_accuracy": 0.9049409031867981,
|
| 466 |
+
"eval_loss": 0.641972541809082,
|
| 467 |
+
"eval_runtime": 487.5431,
|
| 468 |
+
"eval_samples_per_second": 120.594,
|
| 469 |
+
"eval_steps_per_second": 1.885,
|
| 470 |
+
"step": 22000
|
| 471 |
+
}
|
| 472 |
+
],
|
| 473 |
+
"max_steps": 25910,
|
| 474 |
+
"num_train_epochs": 5,
|
| 475 |
+
"total_flos": 1.8496468450847293e+18,
|
| 476 |
+
"trial_name": null,
|
| 477 |
+
"trial_params": null
|
| 478 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4bb41da29b2513ffa40049809d02a074e0d76156f778742aff9ed3611f34b5e4
|
| 3 |
+
size 3631
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|