Upload 7 files
Browse files- config.json +27 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +232 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "roberta-base",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"RobertaForMaskedLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 3072,
|
| 15 |
+
"layer_norm_eps": 1e-05,
|
| 16 |
+
"max_position_embeddings": 514,
|
| 17 |
+
"model_type": "roberta",
|
| 18 |
+
"num_attention_heads": 12,
|
| 19 |
+
"num_hidden_layers": 12,
|
| 20 |
+
"pad_token_id": 1,
|
| 21 |
+
"position_embedding_type": "absolute",
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.30.2",
|
| 24 |
+
"type_vocab_size": 1,
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 50265
|
| 27 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1e740b9e7e59362b2a609bf1c1c007e58a32c6e8d920c13e49379258804ef7f0
|
| 3 |
+
size 997697861
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:855c04b8663e65808228659883427dc08147bf0359a701a1033e41c42eebc4de
|
| 3 |
+
size 498863417
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ba44a916a3c353d8ae915ba6b6f90fef7703aa3303df034a1947dbdcc52e3e09
|
| 3 |
+
size 14575
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7832ff9c490fc2e7a8618936d6f57676cde91cf0724d9071b2daa20d1db2fe37
|
| 3 |
+
size 627
|
trainer_state.json
ADDED
|
@@ -0,0 +1,232 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 2.273698555654846e-06,
|
| 3 |
+
"best_model_checkpoint": "data/further_pretraining/roberta-base/checkpoint-11695",
|
| 4 |
+
"epoch": 6.0,
|
| 5 |
+
"global_step": 14034,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.21,
|
| 12 |
+
"learning_rate": 1.9572466866182133e-05,
|
| 13 |
+
"loss": 0.0508,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.43,
|
| 18 |
+
"learning_rate": 1.914493373236426e-05,
|
| 19 |
+
"loss": 0.0326,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.64,
|
| 24 |
+
"learning_rate": 1.8717400598546387e-05,
|
| 25 |
+
"loss": 0.0135,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.86,
|
| 30 |
+
"learning_rate": 1.828986746472852e-05,
|
| 31 |
+
"loss": 0.0023,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.0,
|
| 36 |
+
"eval_loss": 0.0002380974474363029,
|
| 37 |
+
"eval_runtime": 225.964,
|
| 38 |
+
"eval_samples_per_second": 331.212,
|
| 39 |
+
"eval_steps_per_second": 10.351,
|
| 40 |
+
"step": 2339
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 1.07,
|
| 44 |
+
"learning_rate": 1.7862334330910646e-05,
|
| 45 |
+
"loss": 0.0009,
|
| 46 |
+
"step": 2500
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.28,
|
| 50 |
+
"learning_rate": 1.7434801197092777e-05,
|
| 51 |
+
"loss": 0.0004,
|
| 52 |
+
"step": 3000
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.5,
|
| 56 |
+
"learning_rate": 1.7007268063274904e-05,
|
| 57 |
+
"loss": 0.0004,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 1.71,
|
| 62 |
+
"learning_rate": 1.6579734929457035e-05,
|
| 63 |
+
"loss": 0.0004,
|
| 64 |
+
"step": 4000
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 1.92,
|
| 68 |
+
"learning_rate": 1.6152201795639163e-05,
|
| 69 |
+
"loss": 0.0003,
|
| 70 |
+
"step": 4500
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 2.0,
|
| 74 |
+
"eval_loss": 2.140663673344534e-05,
|
| 75 |
+
"eval_runtime": 228.247,
|
| 76 |
+
"eval_samples_per_second": 327.899,
|
| 77 |
+
"eval_steps_per_second": 10.248,
|
| 78 |
+
"step": 4678
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 2.14,
|
| 82 |
+
"learning_rate": 1.5724668661821293e-05,
|
| 83 |
+
"loss": 0.0001,
|
| 84 |
+
"step": 5000
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 2.35,
|
| 88 |
+
"learning_rate": 1.529713552800342e-05,
|
| 89 |
+
"loss": 0.0001,
|
| 90 |
+
"step": 5500
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 2.57,
|
| 94 |
+
"learning_rate": 1.4869602394185552e-05,
|
| 95 |
+
"loss": 0.0002,
|
| 96 |
+
"step": 6000
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 2.78,
|
| 100 |
+
"learning_rate": 1.444206926036768e-05,
|
| 101 |
+
"loss": 0.0001,
|
| 102 |
+
"step": 6500
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 2.99,
|
| 106 |
+
"learning_rate": 1.4014536126549808e-05,
|
| 107 |
+
"loss": 0.0001,
|
| 108 |
+
"step": 7000
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 3.0,
|
| 112 |
+
"eval_loss": 5.5228130804607645e-05,
|
| 113 |
+
"eval_runtime": 228.2524,
|
| 114 |
+
"eval_samples_per_second": 327.891,
|
| 115 |
+
"eval_steps_per_second": 10.247,
|
| 116 |
+
"step": 7017
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 3.21,
|
| 120 |
+
"learning_rate": 1.3587002992731938e-05,
|
| 121 |
+
"loss": 0.0001,
|
| 122 |
+
"step": 7500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 3.42,
|
| 126 |
+
"learning_rate": 1.3159469858914067e-05,
|
| 127 |
+
"loss": 0.0001,
|
| 128 |
+
"step": 8000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 3.63,
|
| 132 |
+
"learning_rate": 1.2731936725096196e-05,
|
| 133 |
+
"loss": 0.0001,
|
| 134 |
+
"step": 8500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.85,
|
| 138 |
+
"learning_rate": 1.2304403591278325e-05,
|
| 139 |
+
"loss": 0.0004,
|
| 140 |
+
"step": 9000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 4.0,
|
| 144 |
+
"eval_loss": 6.755703907401767e-06,
|
| 145 |
+
"eval_runtime": 228.2467,
|
| 146 |
+
"eval_samples_per_second": 327.9,
|
| 147 |
+
"eval_steps_per_second": 10.248,
|
| 148 |
+
"step": 9356
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 4.06,
|
| 152 |
+
"learning_rate": 1.1876870457460454e-05,
|
| 153 |
+
"loss": 0.0001,
|
| 154 |
+
"step": 9500
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 4.28,
|
| 158 |
+
"learning_rate": 1.1449337323642584e-05,
|
| 159 |
+
"loss": 0.0,
|
| 160 |
+
"step": 10000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 4.49,
|
| 164 |
+
"learning_rate": 1.1021804189824713e-05,
|
| 165 |
+
"loss": 0.0002,
|
| 166 |
+
"step": 10500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 4.7,
|
| 170 |
+
"learning_rate": 1.0594271056006842e-05,
|
| 171 |
+
"loss": 0.0001,
|
| 172 |
+
"step": 11000
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 4.92,
|
| 176 |
+
"learning_rate": 1.016673792218897e-05,
|
| 177 |
+
"loss": 0.0,
|
| 178 |
+
"step": 11500
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 5.0,
|
| 182 |
+
"eval_loss": 2.273698555654846e-06,
|
| 183 |
+
"eval_runtime": 227.9681,
|
| 184 |
+
"eval_samples_per_second": 328.3,
|
| 185 |
+
"eval_steps_per_second": 10.26,
|
| 186 |
+
"step": 11695
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 5.13,
|
| 190 |
+
"learning_rate": 9.739204788371099e-06,
|
| 191 |
+
"loss": 0.0007,
|
| 192 |
+
"step": 12000
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 5.34,
|
| 196 |
+
"learning_rate": 9.311671654553228e-06,
|
| 197 |
+
"loss": 0.0001,
|
| 198 |
+
"step": 12500
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 5.56,
|
| 202 |
+
"learning_rate": 8.884138520735359e-06,
|
| 203 |
+
"loss": 0.0,
|
| 204 |
+
"step": 13000
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 5.77,
|
| 208 |
+
"learning_rate": 8.456605386917486e-06,
|
| 209 |
+
"loss": 0.0001,
|
| 210 |
+
"step": 13500
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 5.99,
|
| 214 |
+
"learning_rate": 8.029072253099615e-06,
|
| 215 |
+
"loss": 0.0001,
|
| 216 |
+
"step": 14000
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 6.0,
|
| 220 |
+
"eval_loss": 3.073917014262406e-06,
|
| 221 |
+
"eval_runtime": 228.2599,
|
| 222 |
+
"eval_samples_per_second": 327.881,
|
| 223 |
+
"eval_steps_per_second": 10.247,
|
| 224 |
+
"step": 14034
|
| 225 |
+
}
|
| 226 |
+
],
|
| 227 |
+
"max_steps": 23390,
|
| 228 |
+
"num_train_epochs": 10,
|
| 229 |
+
"total_flos": 2.955497070599885e+16,
|
| 230 |
+
"trial_name": null,
|
| 231 |
+
"trial_params": null
|
| 232 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ff3541a3dc2878f250bcf18ec3b7906803eb920ac9c34d528d868ee32348435f
|
| 3 |
+
size 3963
|