Upload folder using huggingface_hub
Browse files- config.json +30 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +16 -0
- trainer_state.json +610 -0
- training_args.bin +3 -0
- training_metadata.json +10 -0
config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_cross_attention": false,
|
| 3 |
+
"architectures": [
|
| 4 |
+
"RobertaForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"dtype": "float32",
|
| 10 |
+
"eos_token_id": 2,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout_prob": 0.1,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"is_decoder": false,
|
| 17 |
+
"layer_norm_eps": 1e-05,
|
| 18 |
+
"max_position_embeddings": 514,
|
| 19 |
+
"model_type": "roberta",
|
| 20 |
+
"num_attention_heads": 12,
|
| 21 |
+
"num_hidden_layers": 12,
|
| 22 |
+
"output_past": true,
|
| 23 |
+
"pad_token_id": 1,
|
| 24 |
+
"problem_type": "single_label_classification",
|
| 25 |
+
"tie_word_embeddings": true,
|
| 26 |
+
"transformers_version": "5.0.0",
|
| 27 |
+
"type_vocab_size": 1,
|
| 28 |
+
"use_cache": false,
|
| 29 |
+
"vocab_size": 50265
|
| 30 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:209e18b1560b4629e233cb8f109fc79fe8783636765d1587b501d3d9d6d04632
|
| 3 |
+
size 498612800
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8875c9eda4a9444a1d3d0f08f68b7a72f17ba43003aa9550550c60e286f2078f
|
| 3 |
+
size 997348747
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e35fdb8910d426c9c16878ea3078dc0568d86e073d0a7f891b2f09ffe5b5b22e
|
| 3 |
+
size 14645
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c75de975c68bbeb8ce37282f26021c4e18754c36408856eaa382dc6627e5aaf3
|
| 3 |
+
size 1465
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<s>",
|
| 5 |
+
"cls_token": "<s>",
|
| 6 |
+
"eos_token": "</s>",
|
| 7 |
+
"errors": "replace",
|
| 8 |
+
"is_local": true,
|
| 9 |
+
"mask_token": "<mask>",
|
| 10 |
+
"model_max_length": 512,
|
| 11 |
+
"pad_token": "<pad>",
|
| 12 |
+
"sep_token": "</s>",
|
| 13 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 14 |
+
"trim_offsets": true,
|
| 15 |
+
"unk_token": "<unk>"
|
| 16 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,610 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 4000,
|
| 3 |
+
"best_metric": 0.6931638121604919,
|
| 4 |
+
"best_model_checkpoint": "/content/drive/MyDrive/PolyGuard/model_final/checkpoint-4000",
|
| 5 |
+
"epoch": 2.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 8000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.025,
|
| 14 |
+
"grad_norm": 4.840836524963379,
|
| 15 |
+
"learning_rate": 2.4750000000000002e-05,
|
| 16 |
+
"loss": 0.6476753997802734,
|
| 17 |
+
"step": 100
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.05,
|
| 21 |
+
"grad_norm": 3.3573901653289795,
|
| 22 |
+
"learning_rate": 4.975e-05,
|
| 23 |
+
"loss": 0.6003963470458984,
|
| 24 |
+
"step": 200
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.075,
|
| 28 |
+
"grad_norm": 5.3957953453063965,
|
| 29 |
+
"learning_rate": 4.96867088607595e-05,
|
| 30 |
+
"loss": 0.5752050399780273,
|
| 31 |
+
"step": 300
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.1,
|
| 35 |
+
"grad_norm": 6.348972797393799,
|
| 36 |
+
"learning_rate": 4.937025316455696e-05,
|
| 37 |
+
"loss": 0.5668778610229492,
|
| 38 |
+
"step": 400
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.125,
|
| 42 |
+
"grad_norm": 3.777247428894043,
|
| 43 |
+
"learning_rate": 4.905379746835443e-05,
|
| 44 |
+
"loss": 0.6448370361328125,
|
| 45 |
+
"step": 500
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.15,
|
| 49 |
+
"grad_norm": 5.974273204803467,
|
| 50 |
+
"learning_rate": 4.87373417721519e-05,
|
| 51 |
+
"loss": 0.5914559173583984,
|
| 52 |
+
"step": 600
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.175,
|
| 56 |
+
"grad_norm": 5.3988165855407715,
|
| 57 |
+
"learning_rate": 4.842088607594937e-05,
|
| 58 |
+
"loss": 0.6363700103759765,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.2,
|
| 63 |
+
"grad_norm": 2.299058675765991,
|
| 64 |
+
"learning_rate": 4.810443037974684e-05,
|
| 65 |
+
"loss": 0.5667419815063477,
|
| 66 |
+
"step": 800
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.225,
|
| 70 |
+
"grad_norm": 3.1783978939056396,
|
| 71 |
+
"learning_rate": 4.7787974683544305e-05,
|
| 72 |
+
"loss": 0.5284980392456055,
|
| 73 |
+
"step": 900
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.25,
|
| 77 |
+
"grad_norm": 3.9692907333374023,
|
| 78 |
+
"learning_rate": 4.747151898734177e-05,
|
| 79 |
+
"loss": 0.56060546875,
|
| 80 |
+
"step": 1000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.275,
|
| 84 |
+
"grad_norm": 2.46657657623291,
|
| 85 |
+
"learning_rate": 4.715506329113925e-05,
|
| 86 |
+
"loss": 0.5778974151611328,
|
| 87 |
+
"step": 1100
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.3,
|
| 91 |
+
"grad_norm": 1.4161385297775269,
|
| 92 |
+
"learning_rate": 4.683860759493671e-05,
|
| 93 |
+
"loss": 0.5550444412231446,
|
| 94 |
+
"step": 1200
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.325,
|
| 98 |
+
"grad_norm": 6.127955436706543,
|
| 99 |
+
"learning_rate": 4.652215189873418e-05,
|
| 100 |
+
"loss": 0.638050308227539,
|
| 101 |
+
"step": 1300
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.35,
|
| 105 |
+
"grad_norm": 2.9411966800689697,
|
| 106 |
+
"learning_rate": 4.620569620253164e-05,
|
| 107 |
+
"loss": 0.5464860153198242,
|
| 108 |
+
"step": 1400
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.375,
|
| 112 |
+
"grad_norm": 5.668337345123291,
|
| 113 |
+
"learning_rate": 4.588924050632912e-05,
|
| 114 |
+
"loss": 0.6324460220336914,
|
| 115 |
+
"step": 1500
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.4,
|
| 119 |
+
"grad_norm": 5.304285049438477,
|
| 120 |
+
"learning_rate": 4.5572784810126585e-05,
|
| 121 |
+
"loss": 0.5795536041259766,
|
| 122 |
+
"step": 1600
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.425,
|
| 126 |
+
"grad_norm": 1.6371102333068848,
|
| 127 |
+
"learning_rate": 4.525632911392405e-05,
|
| 128 |
+
"loss": 0.5774750137329101,
|
| 129 |
+
"step": 1700
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.45,
|
| 133 |
+
"grad_norm": 1.4340555667877197,
|
| 134 |
+
"learning_rate": 4.493987341772152e-05,
|
| 135 |
+
"loss": 0.6004017257690429,
|
| 136 |
+
"step": 1800
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.475,
|
| 140 |
+
"grad_norm": 2.997642755508423,
|
| 141 |
+
"learning_rate": 4.462341772151899e-05,
|
| 142 |
+
"loss": 0.6037093734741211,
|
| 143 |
+
"step": 1900
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.5,
|
| 147 |
+
"grad_norm": 1.124809741973877,
|
| 148 |
+
"learning_rate": 4.430696202531646e-05,
|
| 149 |
+
"loss": 0.5478248596191406,
|
| 150 |
+
"step": 2000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.525,
|
| 154 |
+
"grad_norm": 1.4012881517410278,
|
| 155 |
+
"learning_rate": 4.399050632911393e-05,
|
| 156 |
+
"loss": 0.5746703720092774,
|
| 157 |
+
"step": 2100
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.55,
|
| 161 |
+
"grad_norm": 2.774062395095825,
|
| 162 |
+
"learning_rate": 4.367405063291139e-05,
|
| 163 |
+
"loss": 0.5518299865722657,
|
| 164 |
+
"step": 2200
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.575,
|
| 168 |
+
"grad_norm": 0.9460004568099976,
|
| 169 |
+
"learning_rate": 4.3357594936708864e-05,
|
| 170 |
+
"loss": 0.5653076553344727,
|
| 171 |
+
"step": 2300
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.6,
|
| 175 |
+
"grad_norm": 4.564599514007568,
|
| 176 |
+
"learning_rate": 4.304113924050633e-05,
|
| 177 |
+
"loss": 0.626568832397461,
|
| 178 |
+
"step": 2400
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.625,
|
| 182 |
+
"grad_norm": 0.9103949666023254,
|
| 183 |
+
"learning_rate": 4.27246835443038e-05,
|
| 184 |
+
"loss": 0.5939551544189453,
|
| 185 |
+
"step": 2500
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.65,
|
| 189 |
+
"grad_norm": 3.449150800704956,
|
| 190 |
+
"learning_rate": 4.2408227848101265e-05,
|
| 191 |
+
"loss": 0.5971554946899414,
|
| 192 |
+
"step": 2600
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.675,
|
| 196 |
+
"grad_norm": 3.5549769401550293,
|
| 197 |
+
"learning_rate": 4.2091772151898736e-05,
|
| 198 |
+
"loss": 0.5751391983032227,
|
| 199 |
+
"step": 2700
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.7,
|
| 203 |
+
"grad_norm": 1.6542292833328247,
|
| 204 |
+
"learning_rate": 4.177531645569621e-05,
|
| 205 |
+
"loss": 0.5986330032348632,
|
| 206 |
+
"step": 2800
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.725,
|
| 210 |
+
"grad_norm": 1.3074430227279663,
|
| 211 |
+
"learning_rate": 4.145886075949367e-05,
|
| 212 |
+
"loss": 0.5717515182495118,
|
| 213 |
+
"step": 2900
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.75,
|
| 217 |
+
"grad_norm": 3.095973014831543,
|
| 218 |
+
"learning_rate": 4.114240506329114e-05,
|
| 219 |
+
"loss": 0.5813043594360352,
|
| 220 |
+
"step": 3000
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.775,
|
| 224 |
+
"grad_norm": 2.611328363418579,
|
| 225 |
+
"learning_rate": 4.0825949367088615e-05,
|
| 226 |
+
"loss": 0.6051469039916992,
|
| 227 |
+
"step": 3100
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.8,
|
| 231 |
+
"grad_norm": 2.4239795207977295,
|
| 232 |
+
"learning_rate": 4.050949367088608e-05,
|
| 233 |
+
"loss": 0.5933720779418945,
|
| 234 |
+
"step": 3200
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.825,
|
| 238 |
+
"grad_norm": 1.3653069734573364,
|
| 239 |
+
"learning_rate": 4.0193037974683544e-05,
|
| 240 |
+
"loss": 0.5751293182373047,
|
| 241 |
+
"step": 3300
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.85,
|
| 245 |
+
"grad_norm": 1.4362722635269165,
|
| 246 |
+
"learning_rate": 3.9876582278481015e-05,
|
| 247 |
+
"loss": 0.5673767852783204,
|
| 248 |
+
"step": 3400
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.875,
|
| 252 |
+
"grad_norm": 1.1265358924865723,
|
| 253 |
+
"learning_rate": 3.956012658227849e-05,
|
| 254 |
+
"loss": 0.6048561477661133,
|
| 255 |
+
"step": 3500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.9,
|
| 259 |
+
"grad_norm": 1.3712810277938843,
|
| 260 |
+
"learning_rate": 3.924367088607595e-05,
|
| 261 |
+
"loss": 0.6653845977783203,
|
| 262 |
+
"step": 3600
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.925,
|
| 266 |
+
"grad_norm": 1.213412880897522,
|
| 267 |
+
"learning_rate": 3.8927215189873416e-05,
|
| 268 |
+
"loss": 0.6943452453613281,
|
| 269 |
+
"step": 3700
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.95,
|
| 273 |
+
"grad_norm": 2.019483804702759,
|
| 274 |
+
"learning_rate": 3.861075949367089e-05,
|
| 275 |
+
"loss": 0.6974296569824219,
|
| 276 |
+
"step": 3800
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.975,
|
| 280 |
+
"grad_norm": 1.273471713066101,
|
| 281 |
+
"learning_rate": 3.829430379746836e-05,
|
| 282 |
+
"loss": 0.706220474243164,
|
| 283 |
+
"step": 3900
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.0,
|
| 287 |
+
"grad_norm": 2.3656868934631348,
|
| 288 |
+
"learning_rate": 3.7977848101265823e-05,
|
| 289 |
+
"loss": 0.702726058959961,
|
| 290 |
+
"step": 4000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.0,
|
| 294 |
+
"eval_loss": 0.6931638121604919,
|
| 295 |
+
"eval_runtime": 166.9577,
|
| 296 |
+
"eval_samples_per_second": 47.916,
|
| 297 |
+
"eval_steps_per_second": 5.99,
|
| 298 |
+
"step": 4000
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 1.025,
|
| 302 |
+
"grad_norm": 1.4108144044876099,
|
| 303 |
+
"learning_rate": 3.7661392405063295e-05,
|
| 304 |
+
"loss": 0.6999430847167969,
|
| 305 |
+
"step": 4100
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 1.05,
|
| 309 |
+
"grad_norm": 2.259119987487793,
|
| 310 |
+
"learning_rate": 3.734493670886076e-05,
|
| 311 |
+
"loss": 0.7005876922607421,
|
| 312 |
+
"step": 4200
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 1.075,
|
| 316 |
+
"grad_norm": 0.8190616965293884,
|
| 317 |
+
"learning_rate": 3.702848101265823e-05,
|
| 318 |
+
"loss": 0.6983786010742188,
|
| 319 |
+
"step": 4300
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 1.1,
|
| 323 |
+
"grad_norm": 1.3949053287506104,
|
| 324 |
+
"learning_rate": 3.67120253164557e-05,
|
| 325 |
+
"loss": 0.6970309448242188,
|
| 326 |
+
"step": 4400
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 1.125,
|
| 330 |
+
"grad_norm": 1.9007196426391602,
|
| 331 |
+
"learning_rate": 3.639556962025317e-05,
|
| 332 |
+
"loss": 0.7024803161621094,
|
| 333 |
+
"step": 4500
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 1.15,
|
| 337 |
+
"grad_norm": 3.2963850498199463,
|
| 338 |
+
"learning_rate": 3.607911392405063e-05,
|
| 339 |
+
"loss": 0.6969153594970703,
|
| 340 |
+
"step": 4600
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 1.175,
|
| 344 |
+
"grad_norm": 2.8197951316833496,
|
| 345 |
+
"learning_rate": 3.57626582278481e-05,
|
| 346 |
+
"loss": 0.6913418579101562,
|
| 347 |
+
"step": 4700
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 1.2,
|
| 351 |
+
"grad_norm": 3.7260422706604004,
|
| 352 |
+
"learning_rate": 3.5446202531645574e-05,
|
| 353 |
+
"loss": 0.6954251098632812,
|
| 354 |
+
"step": 4800
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.225,
|
| 358 |
+
"grad_norm": 2.538835048675537,
|
| 359 |
+
"learning_rate": 3.512974683544304e-05,
|
| 360 |
+
"loss": 0.7016423797607422,
|
| 361 |
+
"step": 4900
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 1.25,
|
| 365 |
+
"grad_norm": 1.9548262357711792,
|
| 366 |
+
"learning_rate": 3.48132911392405e-05,
|
| 367 |
+
"loss": 0.6976743316650391,
|
| 368 |
+
"step": 5000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 1.275,
|
| 372 |
+
"grad_norm": 4.0937180519104,
|
| 373 |
+
"learning_rate": 3.4496835443037975e-05,
|
| 374 |
+
"loss": 0.700084228515625,
|
| 375 |
+
"step": 5100
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 1.3,
|
| 379 |
+
"grad_norm": 2.521569013595581,
|
| 380 |
+
"learning_rate": 3.4180379746835446e-05,
|
| 381 |
+
"loss": 0.6976382446289062,
|
| 382 |
+
"step": 5200
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 1.325,
|
| 386 |
+
"grad_norm": 4.047502517700195,
|
| 387 |
+
"learning_rate": 3.386392405063291e-05,
|
| 388 |
+
"loss": 0.6967656707763672,
|
| 389 |
+
"step": 5300
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.35,
|
| 393 |
+
"grad_norm": 3.4342639446258545,
|
| 394 |
+
"learning_rate": 3.354746835443038e-05,
|
| 395 |
+
"loss": 0.6955546569824219,
|
| 396 |
+
"step": 5400
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 1.375,
|
| 400 |
+
"grad_norm": 1.5393496751785278,
|
| 401 |
+
"learning_rate": 3.3231012658227854e-05,
|
| 402 |
+
"loss": 0.6965518951416015,
|
| 403 |
+
"step": 5500
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 1.4,
|
| 407 |
+
"grad_norm": 4.321380138397217,
|
| 408 |
+
"learning_rate": 3.291455696202532e-05,
|
| 409 |
+
"loss": 0.6958496856689453,
|
| 410 |
+
"step": 5600
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 1.425,
|
| 414 |
+
"grad_norm": 1.1191785335540771,
|
| 415 |
+
"learning_rate": 3.259810126582279e-05,
|
| 416 |
+
"loss": 0.6999098968505859,
|
| 417 |
+
"step": 5700
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 1.45,
|
| 421 |
+
"grad_norm": 0.8614036440849304,
|
| 422 |
+
"learning_rate": 3.2281645569620254e-05,
|
| 423 |
+
"loss": 0.6959712219238281,
|
| 424 |
+
"step": 5800
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 1.475,
|
| 428 |
+
"grad_norm": 0.9664958715438843,
|
| 429 |
+
"learning_rate": 3.1965189873417725e-05,
|
| 430 |
+
"loss": 0.6925695037841797,
|
| 431 |
+
"step": 5900
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 1.5,
|
| 435 |
+
"grad_norm": 2.1453211307525635,
|
| 436 |
+
"learning_rate": 3.164873417721519e-05,
|
| 437 |
+
"loss": 0.698175048828125,
|
| 438 |
+
"step": 6000
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 1.525,
|
| 442 |
+
"grad_norm": 2.1440930366516113,
|
| 443 |
+
"learning_rate": 3.133227848101266e-05,
|
| 444 |
+
"loss": 0.682925796508789,
|
| 445 |
+
"step": 6100
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 1.55,
|
| 449 |
+
"grad_norm": 0.8990124464035034,
|
| 450 |
+
"learning_rate": 3.1015822784810126e-05,
|
| 451 |
+
"loss": 0.6517456817626953,
|
| 452 |
+
"step": 6200
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 1.575,
|
| 456 |
+
"grad_norm": 1.3929342031478882,
|
| 457 |
+
"learning_rate": 3.06993670886076e-05,
|
| 458 |
+
"loss": 0.7006549835205078,
|
| 459 |
+
"step": 6300
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 1.6,
|
| 463 |
+
"grad_norm": 2.1912875175476074,
|
| 464 |
+
"learning_rate": 3.0382911392405065e-05,
|
| 465 |
+
"loss": 0.6975661468505859,
|
| 466 |
+
"step": 6400
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 1.625,
|
| 470 |
+
"grad_norm": 1.1694247722625732,
|
| 471 |
+
"learning_rate": 3.0066455696202533e-05,
|
| 472 |
+
"loss": 0.6955360412597656,
|
| 473 |
+
"step": 6500
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 1.65,
|
| 477 |
+
"grad_norm": 3.340589761734009,
|
| 478 |
+
"learning_rate": 2.975e-05,
|
| 479 |
+
"loss": 0.7008393859863281,
|
| 480 |
+
"step": 6600
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 1.675,
|
| 484 |
+
"grad_norm": 2.42465877532959,
|
| 485 |
+
"learning_rate": 2.9433544303797473e-05,
|
| 486 |
+
"loss": 0.6998232269287109,
|
| 487 |
+
"step": 6700
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 1.7,
|
| 491 |
+
"grad_norm": 1.9139105081558228,
|
| 492 |
+
"learning_rate": 2.9117088607594937e-05,
|
| 493 |
+
"loss": 0.6989698028564453,
|
| 494 |
+
"step": 6800
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 1.725,
|
| 498 |
+
"grad_norm": 0.7264005541801453,
|
| 499 |
+
"learning_rate": 2.8800632911392405e-05,
|
| 500 |
+
"loss": 0.6944959259033203,
|
| 501 |
+
"step": 6900
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 1.75,
|
| 505 |
+
"grad_norm": 1.1322827339172363,
|
| 506 |
+
"learning_rate": 2.8484177215189873e-05,
|
| 507 |
+
"loss": 0.7098442840576172,
|
| 508 |
+
"step": 7000
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 1.775,
|
| 512 |
+
"grad_norm": 2.150141477584839,
|
| 513 |
+
"learning_rate": 2.8167721518987345e-05,
|
| 514 |
+
"loss": 0.6946941375732422,
|
| 515 |
+
"step": 7100
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 1.8,
|
| 519 |
+
"grad_norm": 1.850095510482788,
|
| 520 |
+
"learning_rate": 2.785126582278481e-05,
|
| 521 |
+
"loss": 0.6941130065917969,
|
| 522 |
+
"step": 7200
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 1.825,
|
| 526 |
+
"grad_norm": 1.994320273399353,
|
| 527 |
+
"learning_rate": 2.7534810126582277e-05,
|
| 528 |
+
"loss": 0.6933687591552734,
|
| 529 |
+
"step": 7300
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 1.85,
|
| 533 |
+
"grad_norm": 1.1272798776626587,
|
| 534 |
+
"learning_rate": 2.721835443037975e-05,
|
| 535 |
+
"loss": 0.7010916900634766,
|
| 536 |
+
"step": 7400
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 1.875,
|
| 540 |
+
"grad_norm": 2.2662463188171387,
|
| 541 |
+
"learning_rate": 2.6901898734177217e-05,
|
| 542 |
+
"loss": 0.6951226806640625,
|
| 543 |
+
"step": 7500
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 1.9,
|
| 547 |
+
"grad_norm": 2.8019468784332275,
|
| 548 |
+
"learning_rate": 2.6585443037974685e-05,
|
| 549 |
+
"loss": 0.6966998291015625,
|
| 550 |
+
"step": 7600
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 1.925,
|
| 554 |
+
"grad_norm": 2.3949637413024902,
|
| 555 |
+
"learning_rate": 2.6268987341772156e-05,
|
| 556 |
+
"loss": 0.6956380462646484,
|
| 557 |
+
"step": 7700
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 1.95,
|
| 561 |
+
"grad_norm": 2.5100715160369873,
|
| 562 |
+
"learning_rate": 2.595253164556962e-05,
|
| 563 |
+
"loss": 0.6968719482421875,
|
| 564 |
+
"step": 7800
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 1.975,
|
| 568 |
+
"grad_norm": 6.460758209228516,
|
| 569 |
+
"learning_rate": 2.563607594936709e-05,
|
| 570 |
+
"loss": 0.692848892211914,
|
| 571 |
+
"step": 7900
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 2.0,
|
| 575 |
+
"grad_norm": 3.2004637718200684,
|
| 576 |
+
"learning_rate": 2.5319620253164557e-05,
|
| 577 |
+
"loss": 0.6972612762451171,
|
| 578 |
+
"step": 8000
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 2.0,
|
| 582 |
+
"eval_loss": 0.6932027339935303,
|
| 583 |
+
"eval_runtime": 166.3088,
|
| 584 |
+
"eval_samples_per_second": 48.103,
|
| 585 |
+
"eval_steps_per_second": 6.013,
|
| 586 |
+
"step": 8000
|
| 587 |
+
}
|
| 588 |
+
],
|
| 589 |
+
"logging_steps": 100,
|
| 590 |
+
"max_steps": 16000,
|
| 591 |
+
"num_input_tokens_seen": 0,
|
| 592 |
+
"num_train_epochs": 4,
|
| 593 |
+
"save_steps": 500,
|
| 594 |
+
"stateful_callbacks": {
|
| 595 |
+
"TrainerControl": {
|
| 596 |
+
"args": {
|
| 597 |
+
"should_epoch_stop": false,
|
| 598 |
+
"should_evaluate": false,
|
| 599 |
+
"should_log": false,
|
| 600 |
+
"should_save": true,
|
| 601 |
+
"should_training_stop": false
|
| 602 |
+
},
|
| 603 |
+
"attributes": {}
|
| 604 |
+
}
|
| 605 |
+
},
|
| 606 |
+
"total_flos": 8419553771520000.0,
|
| 607 |
+
"train_batch_size": 8,
|
| 608 |
+
"trial_name": null,
|
| 609 |
+
"trial_params": null
|
| 610 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:144d97010bcda68f4d4b7151c2bcd2b1d3638f1a60032525b2ff2fdfd41fd895
|
| 3 |
+
size 5201
|
training_metadata.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"trained_at": "2026-04-29T00:06:05.040980",
|
| 3 |
+
"base_model": "microsoft/codebert-base",
|
| 4 |
+
"train_samples": 16681,
|
| 5 |
+
"val_samples": 2275,
|
| 6 |
+
"best_f1": 0.6698,
|
| 7 |
+
"epochs_trained": 14,
|
| 8 |
+
"augmented": true,
|
| 9 |
+
"version": "v5_extended"
|
| 10 |
+
}
|