Delete checkpoint-765
Browse files
checkpoint-765/config.json
DELETED
|
@@ -1,43 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"architectures": [
|
| 3 |
-
"DebertaV2ForSequenceClassification"
|
| 4 |
-
],
|
| 5 |
-
"attention_probs_dropout_prob": 0.1,
|
| 6 |
-
"hidden_act": "gelu",
|
| 7 |
-
"hidden_dropout_prob": 0.1,
|
| 8 |
-
"hidden_size": 768,
|
| 9 |
-
"id2label": {
|
| 10 |
-
"0": "\u0e01\u0e32\u0e23\u0e08\u0e49\u0e32\u0e07\u0e23\u0e16\u0e23\u0e31\u0e1a\u0e2a\u0e48\u0e07\u0e2a\u0e34\u0e19\u0e04\u0e49\u0e32",
|
| 11 |
-
"1": "\u0e44\u0e21\u0e48\u0e43\u0e0a\u0e48\u0e01\u0e32\u0e23\u0e08\u0e49\u0e32\u0e07\u0e23\u0e16\u0e23\u0e31\u0e1a\u0e2a\u0e48\u0e07\u0e2a\u0e34\u0e19\u0e04\u0e49\u0e32"
|
| 12 |
-
},
|
| 13 |
-
"initializer_range": 0.02,
|
| 14 |
-
"intermediate_size": 3072,
|
| 15 |
-
"label2id": {
|
| 16 |
-
"entailment": 0,
|
| 17 |
-
"not_entailment": 1
|
| 18 |
-
},
|
| 19 |
-
"layer_norm_eps": 1e-07,
|
| 20 |
-
"legacy": true,
|
| 21 |
-
"max_position_embeddings": 512,
|
| 22 |
-
"max_relative_positions": -1,
|
| 23 |
-
"model_type": "deberta-v2",
|
| 24 |
-
"norm_rel_ebd": "layer_norm",
|
| 25 |
-
"num_attention_heads": 12,
|
| 26 |
-
"num_hidden_layers": 12,
|
| 27 |
-
"pad_token_id": 0,
|
| 28 |
-
"pooler_dropout": 0,
|
| 29 |
-
"pooler_hidden_act": "gelu",
|
| 30 |
-
"pooler_hidden_size": 768,
|
| 31 |
-
"pos_att_type": [
|
| 32 |
-
"p2c",
|
| 33 |
-
"c2p"
|
| 34 |
-
],
|
| 35 |
-
"position_biased_input": false,
|
| 36 |
-
"position_buckets": 256,
|
| 37 |
-
"relative_attention": true,
|
| 38 |
-
"share_att_key": true,
|
| 39 |
-
"torch_dtype": "float32",
|
| 40 |
-
"transformers_version": "4.52.4",
|
| 41 |
-
"type_vocab_size": 0,
|
| 42 |
-
"vocab_size": 251000
|
| 43 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-765/special_tokens_map.json
DELETED
|
@@ -1,51 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"bos_token": {
|
| 3 |
-
"content": "[CLS]",
|
| 4 |
-
"lstrip": false,
|
| 5 |
-
"normalized": false,
|
| 6 |
-
"rstrip": false,
|
| 7 |
-
"single_word": false
|
| 8 |
-
},
|
| 9 |
-
"cls_token": {
|
| 10 |
-
"content": "[CLS]",
|
| 11 |
-
"lstrip": false,
|
| 12 |
-
"normalized": false,
|
| 13 |
-
"rstrip": false,
|
| 14 |
-
"single_word": false
|
| 15 |
-
},
|
| 16 |
-
"eos_token": {
|
| 17 |
-
"content": "[SEP]",
|
| 18 |
-
"lstrip": false,
|
| 19 |
-
"normalized": false,
|
| 20 |
-
"rstrip": false,
|
| 21 |
-
"single_word": false
|
| 22 |
-
},
|
| 23 |
-
"mask_token": {
|
| 24 |
-
"content": "[MASK]",
|
| 25 |
-
"lstrip": false,
|
| 26 |
-
"normalized": false,
|
| 27 |
-
"rstrip": false,
|
| 28 |
-
"single_word": false
|
| 29 |
-
},
|
| 30 |
-
"pad_token": {
|
| 31 |
-
"content": "[PAD]",
|
| 32 |
-
"lstrip": false,
|
| 33 |
-
"normalized": false,
|
| 34 |
-
"rstrip": false,
|
| 35 |
-
"single_word": false
|
| 36 |
-
},
|
| 37 |
-
"sep_token": {
|
| 38 |
-
"content": "[SEP]",
|
| 39 |
-
"lstrip": false,
|
| 40 |
-
"normalized": false,
|
| 41 |
-
"rstrip": false,
|
| 42 |
-
"single_word": false
|
| 43 |
-
},
|
| 44 |
-
"unk_token": {
|
| 45 |
-
"content": "[UNK]",
|
| 46 |
-
"lstrip": false,
|
| 47 |
-
"normalized": true,
|
| 48 |
-
"rstrip": false,
|
| 49 |
-
"single_word": false
|
| 50 |
-
}
|
| 51 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-765/tokenizer.json
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:824e7d82bc5560a30dcf36d84d4598ddcb0ca3c1b3fe38e698105f778bf89d3e
|
| 3 |
-
size 16316224
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-765/tokenizer_config.json
DELETED
|
@@ -1,59 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"added_tokens_decoder": {
|
| 3 |
-
"0": {
|
| 4 |
-
"content": "[PAD]",
|
| 5 |
-
"lstrip": false,
|
| 6 |
-
"normalized": false,
|
| 7 |
-
"rstrip": false,
|
| 8 |
-
"single_word": false,
|
| 9 |
-
"special": true
|
| 10 |
-
},
|
| 11 |
-
"1": {
|
| 12 |
-
"content": "[CLS]",
|
| 13 |
-
"lstrip": false,
|
| 14 |
-
"normalized": false,
|
| 15 |
-
"rstrip": false,
|
| 16 |
-
"single_word": false,
|
| 17 |
-
"special": true
|
| 18 |
-
},
|
| 19 |
-
"2": {
|
| 20 |
-
"content": "[SEP]",
|
| 21 |
-
"lstrip": false,
|
| 22 |
-
"normalized": false,
|
| 23 |
-
"rstrip": false,
|
| 24 |
-
"single_word": false,
|
| 25 |
-
"special": true
|
| 26 |
-
},
|
| 27 |
-
"3": {
|
| 28 |
-
"content": "[UNK]",
|
| 29 |
-
"lstrip": false,
|
| 30 |
-
"normalized": true,
|
| 31 |
-
"rstrip": false,
|
| 32 |
-
"single_word": false,
|
| 33 |
-
"special": true
|
| 34 |
-
},
|
| 35 |
-
"250101": {
|
| 36 |
-
"content": "[MASK]",
|
| 37 |
-
"lstrip": false,
|
| 38 |
-
"normalized": false,
|
| 39 |
-
"rstrip": false,
|
| 40 |
-
"single_word": false,
|
| 41 |
-
"special": true
|
| 42 |
-
}
|
| 43 |
-
},
|
| 44 |
-
"bos_token": "[CLS]",
|
| 45 |
-
"clean_up_tokenization_spaces": true,
|
| 46 |
-
"cls_token": "[CLS]",
|
| 47 |
-
"do_lower_case": false,
|
| 48 |
-
"eos_token": "[SEP]",
|
| 49 |
-
"extra_special_tokens": {},
|
| 50 |
-
"mask_token": "[MASK]",
|
| 51 |
-
"model_max_length": 512,
|
| 52 |
-
"pad_token": "[PAD]",
|
| 53 |
-
"sep_token": "[SEP]",
|
| 54 |
-
"sp_model_kwargs": {},
|
| 55 |
-
"split_by_punct": false,
|
| 56 |
-
"tokenizer_class": "DebertaV2Tokenizer",
|
| 57 |
-
"unk_token": "[UNK]",
|
| 58 |
-
"vocab_type": "spm"
|
| 59 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-765/trainer_state.json
DELETED
|
@@ -1,566 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_global_step": null,
|
| 3 |
-
"best_metric": null,
|
| 4 |
-
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 5.0,
|
| 6 |
-
"eval_steps": 500,
|
| 7 |
-
"global_step": 765,
|
| 8 |
-
"is_hyper_param_search": false,
|
| 9 |
-
"is_local_process_zero": true,
|
| 10 |
-
"is_world_process_zero": true,
|
| 11 |
-
"log_history": [
|
| 12 |
-
{
|
| 13 |
-
"epoch": 0.06535947712418301,
|
| 14 |
-
"grad_norm": 5.113699436187744,
|
| 15 |
-
"learning_rate": 1.9764705882352945e-05,
|
| 16 |
-
"loss": 0.5393,
|
| 17 |
-
"step": 10
|
| 18 |
-
},
|
| 19 |
-
{
|
| 20 |
-
"epoch": 0.13071895424836602,
|
| 21 |
-
"grad_norm": 3.5036208629608154,
|
| 22 |
-
"learning_rate": 1.950326797385621e-05,
|
| 23 |
-
"loss": 0.4357,
|
| 24 |
-
"step": 20
|
| 25 |
-
},
|
| 26 |
-
{
|
| 27 |
-
"epoch": 0.19607843137254902,
|
| 28 |
-
"grad_norm": 3.724426746368408,
|
| 29 |
-
"learning_rate": 1.924183006535948e-05,
|
| 30 |
-
"loss": 0.3842,
|
| 31 |
-
"step": 30
|
| 32 |
-
},
|
| 33 |
-
{
|
| 34 |
-
"epoch": 0.26143790849673204,
|
| 35 |
-
"grad_norm": 4.891366481781006,
|
| 36 |
-
"learning_rate": 1.8980392156862747e-05,
|
| 37 |
-
"loss": 0.2215,
|
| 38 |
-
"step": 40
|
| 39 |
-
},
|
| 40 |
-
{
|
| 41 |
-
"epoch": 0.32679738562091504,
|
| 42 |
-
"grad_norm": 13.632070541381836,
|
| 43 |
-
"learning_rate": 1.8718954248366014e-05,
|
| 44 |
-
"loss": 0.3702,
|
| 45 |
-
"step": 50
|
| 46 |
-
},
|
| 47 |
-
{
|
| 48 |
-
"epoch": 0.39215686274509803,
|
| 49 |
-
"grad_norm": 4.439073085784912,
|
| 50 |
-
"learning_rate": 1.8457516339869284e-05,
|
| 51 |
-
"loss": 0.1825,
|
| 52 |
-
"step": 60
|
| 53 |
-
},
|
| 54 |
-
{
|
| 55 |
-
"epoch": 0.45751633986928103,
|
| 56 |
-
"grad_norm": 4.005460739135742,
|
| 57 |
-
"learning_rate": 1.8196078431372552e-05,
|
| 58 |
-
"loss": 0.164,
|
| 59 |
-
"step": 70
|
| 60 |
-
},
|
| 61 |
-
{
|
| 62 |
-
"epoch": 0.5228758169934641,
|
| 63 |
-
"grad_norm": 1.0837407112121582,
|
| 64 |
-
"learning_rate": 1.793464052287582e-05,
|
| 65 |
-
"loss": 0.2075,
|
| 66 |
-
"step": 80
|
| 67 |
-
},
|
| 68 |
-
{
|
| 69 |
-
"epoch": 0.5882352941176471,
|
| 70 |
-
"grad_norm": 5.353822708129883,
|
| 71 |
-
"learning_rate": 1.7673202614379086e-05,
|
| 72 |
-
"loss": 0.2426,
|
| 73 |
-
"step": 90
|
| 74 |
-
},
|
| 75 |
-
{
|
| 76 |
-
"epoch": 0.6535947712418301,
|
| 77 |
-
"grad_norm": 0.31490930914878845,
|
| 78 |
-
"learning_rate": 1.7411764705882353e-05,
|
| 79 |
-
"loss": 0.128,
|
| 80 |
-
"step": 100
|
| 81 |
-
},
|
| 82 |
-
{
|
| 83 |
-
"epoch": 0.7189542483660131,
|
| 84 |
-
"grad_norm": 0.4094616770744324,
|
| 85 |
-
"learning_rate": 1.715032679738562e-05,
|
| 86 |
-
"loss": 0.1713,
|
| 87 |
-
"step": 110
|
| 88 |
-
},
|
| 89 |
-
{
|
| 90 |
-
"epoch": 0.7843137254901961,
|
| 91 |
-
"grad_norm": 3.661475419998169,
|
| 92 |
-
"learning_rate": 1.688888888888889e-05,
|
| 93 |
-
"loss": 0.131,
|
| 94 |
-
"step": 120
|
| 95 |
-
},
|
| 96 |
-
{
|
| 97 |
-
"epoch": 0.8496732026143791,
|
| 98 |
-
"grad_norm": 0.22965025901794434,
|
| 99 |
-
"learning_rate": 1.662745098039216e-05,
|
| 100 |
-
"loss": 0.2523,
|
| 101 |
-
"step": 130
|
| 102 |
-
},
|
| 103 |
-
{
|
| 104 |
-
"epoch": 0.9150326797385621,
|
| 105 |
-
"grad_norm": 0.2205059677362442,
|
| 106 |
-
"learning_rate": 1.6366013071895426e-05,
|
| 107 |
-
"loss": 0.0297,
|
| 108 |
-
"step": 140
|
| 109 |
-
},
|
| 110 |
-
{
|
| 111 |
-
"epoch": 0.9803921568627451,
|
| 112 |
-
"grad_norm": 0.13449294865131378,
|
| 113 |
-
"learning_rate": 1.6104575163398693e-05,
|
| 114 |
-
"loss": 0.0598,
|
| 115 |
-
"step": 150
|
| 116 |
-
},
|
| 117 |
-
{
|
| 118 |
-
"epoch": 1.0457516339869282,
|
| 119 |
-
"grad_norm": 33.76457977294922,
|
| 120 |
-
"learning_rate": 1.584313725490196e-05,
|
| 121 |
-
"loss": 0.1965,
|
| 122 |
-
"step": 160
|
| 123 |
-
},
|
| 124 |
-
{
|
| 125 |
-
"epoch": 1.1111111111111112,
|
| 126 |
-
"grad_norm": 0.1320052593946457,
|
| 127 |
-
"learning_rate": 1.558169934640523e-05,
|
| 128 |
-
"loss": 0.0211,
|
| 129 |
-
"step": 170
|
| 130 |
-
},
|
| 131 |
-
{
|
| 132 |
-
"epoch": 1.1764705882352942,
|
| 133 |
-
"grad_norm": 5.923645973205566,
|
| 134 |
-
"learning_rate": 1.5320261437908498e-05,
|
| 135 |
-
"loss": 0.0544,
|
| 136 |
-
"step": 180
|
| 137 |
-
},
|
| 138 |
-
{
|
| 139 |
-
"epoch": 1.2418300653594772,
|
| 140 |
-
"grad_norm": 0.13943836092948914,
|
| 141 |
-
"learning_rate": 1.5058823529411765e-05,
|
| 142 |
-
"loss": 0.0058,
|
| 143 |
-
"step": 190
|
| 144 |
-
},
|
| 145 |
-
{
|
| 146 |
-
"epoch": 1.3071895424836601,
|
| 147 |
-
"grad_norm": 0.05259786546230316,
|
| 148 |
-
"learning_rate": 1.4797385620915034e-05,
|
| 149 |
-
"loss": 0.1858,
|
| 150 |
-
"step": 200
|
| 151 |
-
},
|
| 152 |
-
{
|
| 153 |
-
"epoch": 1.3725490196078431,
|
| 154 |
-
"grad_norm": 0.07875791192054749,
|
| 155 |
-
"learning_rate": 1.4535947712418302e-05,
|
| 156 |
-
"loss": 0.0036,
|
| 157 |
-
"step": 210
|
| 158 |
-
},
|
| 159 |
-
{
|
| 160 |
-
"epoch": 1.4379084967320261,
|
| 161 |
-
"grad_norm": 4.741593360900879,
|
| 162 |
-
"learning_rate": 1.427450980392157e-05,
|
| 163 |
-
"loss": 0.2095,
|
| 164 |
-
"step": 220
|
| 165 |
-
},
|
| 166 |
-
{
|
| 167 |
-
"epoch": 1.5032679738562091,
|
| 168 |
-
"grad_norm": 49.31648635864258,
|
| 169 |
-
"learning_rate": 1.4013071895424836e-05,
|
| 170 |
-
"loss": 0.1011,
|
| 171 |
-
"step": 230
|
| 172 |
-
},
|
| 173 |
-
{
|
| 174 |
-
"epoch": 1.5686274509803921,
|
| 175 |
-
"grad_norm": 0.05375028774142265,
|
| 176 |
-
"learning_rate": 1.3751633986928105e-05,
|
| 177 |
-
"loss": 0.0584,
|
| 178 |
-
"step": 240
|
| 179 |
-
},
|
| 180 |
-
{
|
| 181 |
-
"epoch": 1.6339869281045751,
|
| 182 |
-
"grad_norm": 86.06192779541016,
|
| 183 |
-
"learning_rate": 1.3490196078431374e-05,
|
| 184 |
-
"loss": 0.0779,
|
| 185 |
-
"step": 250
|
| 186 |
-
},
|
| 187 |
-
{
|
| 188 |
-
"epoch": 1.6993464052287581,
|
| 189 |
-
"grad_norm": 1.714403510093689,
|
| 190 |
-
"learning_rate": 1.3228758169934641e-05,
|
| 191 |
-
"loss": 0.1763,
|
| 192 |
-
"step": 260
|
| 193 |
-
},
|
| 194 |
-
{
|
| 195 |
-
"epoch": 1.7647058823529411,
|
| 196 |
-
"grad_norm": 0.0629243478178978,
|
| 197 |
-
"learning_rate": 1.296732026143791e-05,
|
| 198 |
-
"loss": 0.1185,
|
| 199 |
-
"step": 270
|
| 200 |
-
},
|
| 201 |
-
{
|
| 202 |
-
"epoch": 1.8300653594771243,
|
| 203 |
-
"grad_norm": 0.060329265892505646,
|
| 204 |
-
"learning_rate": 1.2705882352941177e-05,
|
| 205 |
-
"loss": 0.004,
|
| 206 |
-
"step": 280
|
| 207 |
-
},
|
| 208 |
-
{
|
| 209 |
-
"epoch": 1.8954248366013071,
|
| 210 |
-
"grad_norm": 0.05619505047798157,
|
| 211 |
-
"learning_rate": 1.2444444444444446e-05,
|
| 212 |
-
"loss": 0.088,
|
| 213 |
-
"step": 290
|
| 214 |
-
},
|
| 215 |
-
{
|
| 216 |
-
"epoch": 1.9607843137254903,
|
| 217 |
-
"grad_norm": 0.08919071406126022,
|
| 218 |
-
"learning_rate": 1.2183006535947712e-05,
|
| 219 |
-
"loss": 0.1948,
|
| 220 |
-
"step": 300
|
| 221 |
-
},
|
| 222 |
-
{
|
| 223 |
-
"epoch": 2.026143790849673,
|
| 224 |
-
"grad_norm": 0.5156506896018982,
|
| 225 |
-
"learning_rate": 1.192156862745098e-05,
|
| 226 |
-
"loss": 0.1916,
|
| 227 |
-
"step": 310
|
| 228 |
-
},
|
| 229 |
-
{
|
| 230 |
-
"epoch": 2.0915032679738563,
|
| 231 |
-
"grad_norm": 0.08507811278104782,
|
| 232 |
-
"learning_rate": 1.166013071895425e-05,
|
| 233 |
-
"loss": 0.0036,
|
| 234 |
-
"step": 320
|
| 235 |
-
},
|
| 236 |
-
{
|
| 237 |
-
"epoch": 2.156862745098039,
|
| 238 |
-
"grad_norm": 0.10324525088071823,
|
| 239 |
-
"learning_rate": 1.1398692810457517e-05,
|
| 240 |
-
"loss": 0.078,
|
| 241 |
-
"step": 330
|
| 242 |
-
},
|
| 243 |
-
{
|
| 244 |
-
"epoch": 2.2222222222222223,
|
| 245 |
-
"grad_norm": 0.08859866112470627,
|
| 246 |
-
"learning_rate": 1.1137254901960786e-05,
|
| 247 |
-
"loss": 0.1087,
|
| 248 |
-
"step": 340
|
| 249 |
-
},
|
| 250 |
-
{
|
| 251 |
-
"epoch": 2.287581699346405,
|
| 252 |
-
"grad_norm": 0.2328844666481018,
|
| 253 |
-
"learning_rate": 1.0875816993464051e-05,
|
| 254 |
-
"loss": 0.052,
|
| 255 |
-
"step": 350
|
| 256 |
-
},
|
| 257 |
-
{
|
| 258 |
-
"epoch": 2.3529411764705883,
|
| 259 |
-
"grad_norm": 56.2834587097168,
|
| 260 |
-
"learning_rate": 1.061437908496732e-05,
|
| 261 |
-
"loss": 0.3771,
|
| 262 |
-
"step": 360
|
| 263 |
-
},
|
| 264 |
-
{
|
| 265 |
-
"epoch": 2.418300653594771,
|
| 266 |
-
"grad_norm": 0.15149499475955963,
|
| 267 |
-
"learning_rate": 1.035294117647059e-05,
|
| 268 |
-
"loss": 0.1658,
|
| 269 |
-
"step": 370
|
| 270 |
-
},
|
| 271 |
-
{
|
| 272 |
-
"epoch": 2.4836601307189543,
|
| 273 |
-
"grad_norm": 0.3156740069389343,
|
| 274 |
-
"learning_rate": 1.0091503267973857e-05,
|
| 275 |
-
"loss": 0.0149,
|
| 276 |
-
"step": 380
|
| 277 |
-
},
|
| 278 |
-
{
|
| 279 |
-
"epoch": 2.549019607843137,
|
| 280 |
-
"grad_norm": 0.05973070114850998,
|
| 281 |
-
"learning_rate": 9.830065359477124e-06,
|
| 282 |
-
"loss": 0.0029,
|
| 283 |
-
"step": 390
|
| 284 |
-
},
|
| 285 |
-
{
|
| 286 |
-
"epoch": 2.6143790849673203,
|
| 287 |
-
"grad_norm": 0.0423654243350029,
|
| 288 |
-
"learning_rate": 9.568627450980393e-06,
|
| 289 |
-
"loss": 0.0019,
|
| 290 |
-
"step": 400
|
| 291 |
-
},
|
| 292 |
-
{
|
| 293 |
-
"epoch": 2.6797385620915035,
|
| 294 |
-
"grad_norm": 0.07144688069820404,
|
| 295 |
-
"learning_rate": 9.307189542483662e-06,
|
| 296 |
-
"loss": 0.0837,
|
| 297 |
-
"step": 410
|
| 298 |
-
},
|
| 299 |
-
{
|
| 300 |
-
"epoch": 2.7450980392156863,
|
| 301 |
-
"grad_norm": 7.487165451049805,
|
| 302 |
-
"learning_rate": 9.045751633986929e-06,
|
| 303 |
-
"loss": 0.0744,
|
| 304 |
-
"step": 420
|
| 305 |
-
},
|
| 306 |
-
{
|
| 307 |
-
"epoch": 2.810457516339869,
|
| 308 |
-
"grad_norm": 0.06851428002119064,
|
| 309 |
-
"learning_rate": 8.784313725490196e-06,
|
| 310 |
-
"loss": 0.002,
|
| 311 |
-
"step": 430
|
| 312 |
-
},
|
| 313 |
-
{
|
| 314 |
-
"epoch": 2.8758169934640523,
|
| 315 |
-
"grad_norm": 0.043260689824819565,
|
| 316 |
-
"learning_rate": 8.522875816993465e-06,
|
| 317 |
-
"loss": 0.0326,
|
| 318 |
-
"step": 440
|
| 319 |
-
},
|
| 320 |
-
{
|
| 321 |
-
"epoch": 2.9411764705882355,
|
| 322 |
-
"grad_norm": 0.05410632863640785,
|
| 323 |
-
"learning_rate": 8.261437908496732e-06,
|
| 324 |
-
"loss": 0.074,
|
| 325 |
-
"step": 450
|
| 326 |
-
},
|
| 327 |
-
{
|
| 328 |
-
"epoch": 3.0065359477124183,
|
| 329 |
-
"grad_norm": 0.04789569228887558,
|
| 330 |
-
"learning_rate": 8.000000000000001e-06,
|
| 331 |
-
"loss": 0.089,
|
| 332 |
-
"step": 460
|
| 333 |
-
},
|
| 334 |
-
{
|
| 335 |
-
"epoch": 3.0718954248366015,
|
| 336 |
-
"grad_norm": 0.06222234293818474,
|
| 337 |
-
"learning_rate": 7.738562091503269e-06,
|
| 338 |
-
"loss": 0.0348,
|
| 339 |
-
"step": 470
|
| 340 |
-
},
|
| 341 |
-
{
|
| 342 |
-
"epoch": 3.1372549019607843,
|
| 343 |
-
"grad_norm": 0.038470156490802765,
|
| 344 |
-
"learning_rate": 7.477124183006537e-06,
|
| 345 |
-
"loss": 0.0452,
|
| 346 |
-
"step": 480
|
| 347 |
-
},
|
| 348 |
-
{
|
| 349 |
-
"epoch": 3.2026143790849675,
|
| 350 |
-
"grad_norm": 32.63681411743164,
|
| 351 |
-
"learning_rate": 7.215686274509805e-06,
|
| 352 |
-
"loss": 0.0597,
|
| 353 |
-
"step": 490
|
| 354 |
-
},
|
| 355 |
-
{
|
| 356 |
-
"epoch": 3.2679738562091503,
|
| 357 |
-
"grad_norm": 72.21328735351562,
|
| 358 |
-
"learning_rate": 6.954248366013072e-06,
|
| 359 |
-
"loss": 0.1867,
|
| 360 |
-
"step": 500
|
| 361 |
-
},
|
| 362 |
-
{
|
| 363 |
-
"epoch": 3.3333333333333335,
|
| 364 |
-
"grad_norm": 0.06187112256884575,
|
| 365 |
-
"learning_rate": 6.69281045751634e-06,
|
| 366 |
-
"loss": 0.0019,
|
| 367 |
-
"step": 510
|
| 368 |
-
},
|
| 369 |
-
{
|
| 370 |
-
"epoch": 3.3986928104575163,
|
| 371 |
-
"grad_norm": 0.05040152370929718,
|
| 372 |
-
"learning_rate": 6.431372549019609e-06,
|
| 373 |
-
"loss": 0.0689,
|
| 374 |
-
"step": 520
|
| 375 |
-
},
|
| 376 |
-
{
|
| 377 |
-
"epoch": 3.4640522875816995,
|
| 378 |
-
"grad_norm": 0.03201011195778847,
|
| 379 |
-
"learning_rate": 6.169934640522876e-06,
|
| 380 |
-
"loss": 0.0059,
|
| 381 |
-
"step": 530
|
| 382 |
-
},
|
| 383 |
-
{
|
| 384 |
-
"epoch": 3.5294117647058822,
|
| 385 |
-
"grad_norm": 0.032609112560749054,
|
| 386 |
-
"learning_rate": 5.908496732026144e-06,
|
| 387 |
-
"loss": 0.0715,
|
| 388 |
-
"step": 540
|
| 389 |
-
},
|
| 390 |
-
{
|
| 391 |
-
"epoch": 3.5947712418300655,
|
| 392 |
-
"grad_norm": 2.648466110229492,
|
| 393 |
-
"learning_rate": 5.6470588235294125e-06,
|
| 394 |
-
"loss": 0.0027,
|
| 395 |
-
"step": 550
|
| 396 |
-
},
|
| 397 |
-
{
|
| 398 |
-
"epoch": 3.6601307189542482,
|
| 399 |
-
"grad_norm": 0.04829758033156395,
|
| 400 |
-
"learning_rate": 5.38562091503268e-06,
|
| 401 |
-
"loss": 0.1518,
|
| 402 |
-
"step": 560
|
| 403 |
-
},
|
| 404 |
-
{
|
| 405 |
-
"epoch": 3.7254901960784315,
|
| 406 |
-
"grad_norm": 5.7935380935668945,
|
| 407 |
-
"learning_rate": 5.124183006535948e-06,
|
| 408 |
-
"loss": 0.0033,
|
| 409 |
-
"step": 570
|
| 410 |
-
},
|
| 411 |
-
{
|
| 412 |
-
"epoch": 3.7908496732026142,
|
| 413 |
-
"grad_norm": 0.040050867944955826,
|
| 414 |
-
"learning_rate": 4.862745098039216e-06,
|
| 415 |
-
"loss": 0.0016,
|
| 416 |
-
"step": 580
|
| 417 |
-
},
|
| 418 |
-
{
|
| 419 |
-
"epoch": 3.8562091503267975,
|
| 420 |
-
"grad_norm": 0.07059869915246964,
|
| 421 |
-
"learning_rate": 4.601307189542484e-06,
|
| 422 |
-
"loss": 0.1626,
|
| 423 |
-
"step": 590
|
| 424 |
-
},
|
| 425 |
-
{
|
| 426 |
-
"epoch": 3.9215686274509802,
|
| 427 |
-
"grad_norm": 0.10601856559515,
|
| 428 |
-
"learning_rate": 4.339869281045752e-06,
|
| 429 |
-
"loss": 0.002,
|
| 430 |
-
"step": 600
|
| 431 |
-
},
|
| 432 |
-
{
|
| 433 |
-
"epoch": 3.9869281045751634,
|
| 434 |
-
"grad_norm": 0.05205712467432022,
|
| 435 |
-
"learning_rate": 4.07843137254902e-06,
|
| 436 |
-
"loss": 0.0143,
|
| 437 |
-
"step": 610
|
| 438 |
-
},
|
| 439 |
-
{
|
| 440 |
-
"epoch": 4.052287581699346,
|
| 441 |
-
"grad_norm": 5.935383319854736,
|
| 442 |
-
"learning_rate": 3.816993464052287e-06,
|
| 443 |
-
"loss": 0.0674,
|
| 444 |
-
"step": 620
|
| 445 |
-
},
|
| 446 |
-
{
|
| 447 |
-
"epoch": 4.117647058823529,
|
| 448 |
-
"grad_norm": 0.052745405584573746,
|
| 449 |
-
"learning_rate": 3.555555555555556e-06,
|
| 450 |
-
"loss": 0.0017,
|
| 451 |
-
"step": 630
|
| 452 |
-
},
|
| 453 |
-
{
|
| 454 |
-
"epoch": 4.183006535947713,
|
| 455 |
-
"grad_norm": 0.052255719900131226,
|
| 456 |
-
"learning_rate": 3.2941176470588236e-06,
|
| 457 |
-
"loss": 0.0773,
|
| 458 |
-
"step": 640
|
| 459 |
-
},
|
| 460 |
-
{
|
| 461 |
-
"epoch": 4.248366013071895,
|
| 462 |
-
"grad_norm": 0.06494952738285065,
|
| 463 |
-
"learning_rate": 3.0326797385620917e-06,
|
| 464 |
-
"loss": 0.0417,
|
| 465 |
-
"step": 650
|
| 466 |
-
},
|
| 467 |
-
{
|
| 468 |
-
"epoch": 4.313725490196078,
|
| 469 |
-
"grad_norm": 6.835853099822998,
|
| 470 |
-
"learning_rate": 2.7712418300653598e-06,
|
| 471 |
-
"loss": 0.1278,
|
| 472 |
-
"step": 660
|
| 473 |
-
},
|
| 474 |
-
{
|
| 475 |
-
"epoch": 4.379084967320262,
|
| 476 |
-
"grad_norm": 0.05561675876379013,
|
| 477 |
-
"learning_rate": 2.5098039215686274e-06,
|
| 478 |
-
"loss": 0.0017,
|
| 479 |
-
"step": 670
|
| 480 |
-
},
|
| 481 |
-
{
|
| 482 |
-
"epoch": 4.444444444444445,
|
| 483 |
-
"grad_norm": 0.04858892410993576,
|
| 484 |
-
"learning_rate": 2.2483660130718955e-06,
|
| 485 |
-
"loss": 0.0016,
|
| 486 |
-
"step": 680
|
| 487 |
-
},
|
| 488 |
-
{
|
| 489 |
-
"epoch": 4.509803921568627,
|
| 490 |
-
"grad_norm": 0.04496558755636215,
|
| 491 |
-
"learning_rate": 1.9869281045751636e-06,
|
| 492 |
-
"loss": 0.0017,
|
| 493 |
-
"step": 690
|
| 494 |
-
},
|
| 495 |
-
{
|
| 496 |
-
"epoch": 4.57516339869281,
|
| 497 |
-
"grad_norm": 0.14414705336093903,
|
| 498 |
-
"learning_rate": 1.7254901960784315e-06,
|
| 499 |
-
"loss": 0.0017,
|
| 500 |
-
"step": 700
|
| 501 |
-
},
|
| 502 |
-
{
|
| 503 |
-
"epoch": 4.640522875816993,
|
| 504 |
-
"grad_norm": 0.07616923749446869,
|
| 505 |
-
"learning_rate": 1.4640522875816996e-06,
|
| 506 |
-
"loss": 0.0016,
|
| 507 |
-
"step": 710
|
| 508 |
-
},
|
| 509 |
-
{
|
| 510 |
-
"epoch": 4.705882352941177,
|
| 511 |
-
"grad_norm": 0.05075491592288017,
|
| 512 |
-
"learning_rate": 1.2026143790849675e-06,
|
| 513 |
-
"loss": 0.0022,
|
| 514 |
-
"step": 720
|
| 515 |
-
},
|
| 516 |
-
{
|
| 517 |
-
"epoch": 4.771241830065359,
|
| 518 |
-
"grad_norm": 0.04712093248963356,
|
| 519 |
-
"learning_rate": 9.411764705882353e-07,
|
| 520 |
-
"loss": 0.0857,
|
| 521 |
-
"step": 730
|
| 522 |
-
},
|
| 523 |
-
{
|
| 524 |
-
"epoch": 4.836601307189542,
|
| 525 |
-
"grad_norm": 0.03912083059549332,
|
| 526 |
-
"learning_rate": 6.797385620915032e-07,
|
| 527 |
-
"loss": 0.0597,
|
| 528 |
-
"step": 740
|
| 529 |
-
},
|
| 530 |
-
{
|
| 531 |
-
"epoch": 4.901960784313726,
|
| 532 |
-
"grad_norm": 0.3302587568759918,
|
| 533 |
-
"learning_rate": 4.1830065359477126e-07,
|
| 534 |
-
"loss": 0.0015,
|
| 535 |
-
"step": 750
|
| 536 |
-
},
|
| 537 |
-
{
|
| 538 |
-
"epoch": 4.967320261437909,
|
| 539 |
-
"grad_norm": 0.034191425889730453,
|
| 540 |
-
"learning_rate": 1.5686274509803921e-07,
|
| 541 |
-
"loss": 0.0775,
|
| 542 |
-
"step": 760
|
| 543 |
-
}
|
| 544 |
-
],
|
| 545 |
-
"logging_steps": 10,
|
| 546 |
-
"max_steps": 765,
|
| 547 |
-
"num_input_tokens_seen": 0,
|
| 548 |
-
"num_train_epochs": 5,
|
| 549 |
-
"save_steps": 500,
|
| 550 |
-
"stateful_callbacks": {
|
| 551 |
-
"TrainerControl": {
|
| 552 |
-
"args": {
|
| 553 |
-
"should_epoch_stop": false,
|
| 554 |
-
"should_evaluate": false,
|
| 555 |
-
"should_log": false,
|
| 556 |
-
"should_save": true,
|
| 557 |
-
"should_training_stop": true
|
| 558 |
-
},
|
| 559 |
-
"attributes": {}
|
| 560 |
-
}
|
| 561 |
-
},
|
| 562 |
-
"total_flos": 355053153232392.0,
|
| 563 |
-
"train_batch_size": 8,
|
| 564 |
-
"trial_name": null,
|
| 565 |
-
"trial_params": null
|
| 566 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-765/training_args.bin
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:cc8f550d6d3617db1b6992d1d5c428d1c2201c6e42324b75997f5dd20f4df04e
|
| 3 |
-
size 5649
|
|
|
|
|
|
|
|
|
|
|
|