Upload folder using huggingface_hub
Browse files- head.pt +3 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +270 -0
- training_args.bin +3 -0
head.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ef46b147e1b704240d9934970aceac7d8e4a0e9192e3b685e0578c4e56dd6a9
|
| 3 |
+
size 34182
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ec1dfb52005e7aebc7739b02cfb8380cf9d99bbc8e8bc6d1e09bf7e34652ef02
|
| 3 |
+
size 15177759456
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c0661a56705c08d93cf530af0335dcc8459550d01ee1b08ada6355d2060bb7f6
|
| 3 |
+
size 335989656
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:45700cf67977fb4a6b69973f3768a8e8ca17093407d3ef3843019fef11840b31
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2c8d5b56fb7354e84bd343528e0612a2d3e81ab6cdf2990d2590a2a362b6ebef
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,270 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 15000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.1,
|
| 13 |
+
"grad_norm": 22.972631454467773,
|
| 14 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 15 |
+
"loss": 1.2552,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.2,
|
| 20 |
+
"grad_norm": 46.487945556640625,
|
| 21 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 22 |
+
"loss": 0.4654,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.3,
|
| 27 |
+
"grad_norm": 13.95026683807373,
|
| 28 |
+
"learning_rate": 5e-05,
|
| 29 |
+
"loss": 0.4089,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.4,
|
| 34 |
+
"grad_norm": 5.9401936531066895,
|
| 35 |
+
"learning_rate": 4.983095894354858e-05,
|
| 36 |
+
"loss": 0.367,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.5,
|
| 41 |
+
"grad_norm": 3.8206381797790527,
|
| 42 |
+
"learning_rate": 4.9326121764495596e-05,
|
| 43 |
+
"loss": 0.3275,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.6,
|
| 48 |
+
"grad_norm": 5.384533882141113,
|
| 49 |
+
"learning_rate": 4.849231551964771e-05,
|
| 50 |
+
"loss": 0.3179,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.7,
|
| 55 |
+
"grad_norm": 12.695955276489258,
|
| 56 |
+
"learning_rate": 4.734081600808531e-05,
|
| 57 |
+
"loss": 0.2881,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.8,
|
| 62 |
+
"grad_norm": 0.10525286942720413,
|
| 63 |
+
"learning_rate": 4.588719528532342e-05,
|
| 64 |
+
"loss": 0.2975,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.9,
|
| 69 |
+
"grad_norm": 5.812150001525879,
|
| 70 |
+
"learning_rate": 4.415111107797445e-05,
|
| 71 |
+
"loss": 0.2413,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 1.0,
|
| 76 |
+
"grad_norm": 0.31458476185798645,
|
| 77 |
+
"learning_rate": 4.215604094671835e-05,
|
| 78 |
+
"loss": 0.2518,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 1.0,
|
| 83 |
+
"eval_accuracy": 0.935,
|
| 84 |
+
"eval_loss": 0.29133525490760803,
|
| 85 |
+
"eval_runtime": 226.6205,
|
| 86 |
+
"eval_samples_per_second": 17.651,
|
| 87 |
+
"eval_steps_per_second": 2.206,
|
| 88 |
+
"step": 5000
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.1,
|
| 92 |
+
"grad_norm": 6.748766899108887,
|
| 93 |
+
"learning_rate": 3.9928964792569655e-05,
|
| 94 |
+
"loss": 0.2025,
|
| 95 |
+
"step": 5500
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.2,
|
| 99 |
+
"grad_norm": 0.6717832684516907,
|
| 100 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 101 |
+
"loss": 0.1751,
|
| 102 |
+
"step": 6000
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.3,
|
| 106 |
+
"grad_norm": 0.013602585531771183,
|
| 107 |
+
"learning_rate": 3.490199415097892e-05,
|
| 108 |
+
"loss": 0.1917,
|
| 109 |
+
"step": 6500
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.4,
|
| 113 |
+
"grad_norm": 38.89192199707031,
|
| 114 |
+
"learning_rate": 3.217008081777726e-05,
|
| 115 |
+
"loss": 0.1644,
|
| 116 |
+
"step": 7000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.5,
|
| 120 |
+
"grad_norm": 1.1758809089660645,
|
| 121 |
+
"learning_rate": 2.9341204441673266e-05,
|
| 122 |
+
"loss": 0.1846,
|
| 123 |
+
"step": 7500
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.6,
|
| 127 |
+
"grad_norm": 9.477453204453923e-06,
|
| 128 |
+
"learning_rate": 2.6453620722761896e-05,
|
| 129 |
+
"loss": 0.1546,
|
| 130 |
+
"step": 8000
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.7,
|
| 134 |
+
"grad_norm": 0.899080753326416,
|
| 135 |
+
"learning_rate": 2.3546379277238107e-05,
|
| 136 |
+
"loss": 0.1478,
|
| 137 |
+
"step": 8500
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.8,
|
| 141 |
+
"grad_norm": 4.797894689545501e-06,
|
| 142 |
+
"learning_rate": 2.0658795558326743e-05,
|
| 143 |
+
"loss": 0.1483,
|
| 144 |
+
"step": 9000
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.9,
|
| 148 |
+
"grad_norm": 0.011263616383075714,
|
| 149 |
+
"learning_rate": 1.7829919182222752e-05,
|
| 150 |
+
"loss": 0.165,
|
| 151 |
+
"step": 9500
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 2.0,
|
| 155 |
+
"grad_norm": 0.0016277525573968887,
|
| 156 |
+
"learning_rate": 1.509800584902108e-05,
|
| 157 |
+
"loss": 0.11,
|
| 158 |
+
"step": 10000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 2.0,
|
| 162 |
+
"eval_accuracy": 0.95825,
|
| 163 |
+
"eval_loss": 0.1764126569032669,
|
| 164 |
+
"eval_runtime": 226.3715,
|
| 165 |
+
"eval_samples_per_second": 17.67,
|
| 166 |
+
"eval_steps_per_second": 2.209,
|
| 167 |
+
"step": 10000
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 2.1,
|
| 171 |
+
"grad_norm": 0.00017825818213168532,
|
| 172 |
+
"learning_rate": 1.2500000000000006e-05,
|
| 173 |
+
"loss": 0.0418,
|
| 174 |
+
"step": 10500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 2.2,
|
| 178 |
+
"grad_norm": 0.011653249152004719,
|
| 179 |
+
"learning_rate": 1.0071035207430352e-05,
|
| 180 |
+
"loss": 0.0573,
|
| 181 |
+
"step": 11000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 2.3,
|
| 185 |
+
"grad_norm": 2.5024826300068526e-06,
|
| 186 |
+
"learning_rate": 7.843959053281663e-06,
|
| 187 |
+
"loss": 0.038,
|
| 188 |
+
"step": 11500
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.4,
|
| 192 |
+
"grad_norm": 0.023587645962834358,
|
| 193 |
+
"learning_rate": 5.848888922025553e-06,
|
| 194 |
+
"loss": 0.0434,
|
| 195 |
+
"step": 12000
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.5,
|
| 199 |
+
"grad_norm": 0.05811120569705963,
|
| 200 |
+
"learning_rate": 4.112804714676594e-06,
|
| 201 |
+
"loss": 0.0395,
|
| 202 |
+
"step": 12500
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 2.6,
|
| 206 |
+
"grad_norm": 0.06877656280994415,
|
| 207 |
+
"learning_rate": 2.659183991914696e-06,
|
| 208 |
+
"loss": 0.0387,
|
| 209 |
+
"step": 13000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 2.7,
|
| 213 |
+
"grad_norm": 9.670445251686033e-06,
|
| 214 |
+
"learning_rate": 1.5076844803522922e-06,
|
| 215 |
+
"loss": 0.037,
|
| 216 |
+
"step": 13500
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 2.8,
|
| 220 |
+
"grad_norm": 61.91555404663086,
|
| 221 |
+
"learning_rate": 6.738782355044049e-07,
|
| 222 |
+
"loss": 0.0528,
|
| 223 |
+
"step": 14000
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 2.9,
|
| 227 |
+
"grad_norm": 1.9232193153584376e-05,
|
| 228 |
+
"learning_rate": 1.6904105645142444e-07,
|
| 229 |
+
"loss": 0.0432,
|
| 230 |
+
"step": 14500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 3.0,
|
| 234 |
+
"grad_norm": 1.5172378908800965e-09,
|
| 235 |
+
"learning_rate": 0.0,
|
| 236 |
+
"loss": 0.0381,
|
| 237 |
+
"step": 15000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 3.0,
|
| 241 |
+
"eval_accuracy": 0.959,
|
| 242 |
+
"eval_loss": 0.24073761701583862,
|
| 243 |
+
"eval_runtime": 226.1826,
|
| 244 |
+
"eval_samples_per_second": 17.685,
|
| 245 |
+
"eval_steps_per_second": 2.211,
|
| 246 |
+
"step": 15000
|
| 247 |
+
}
|
| 248 |
+
],
|
| 249 |
+
"logging_steps": 500,
|
| 250 |
+
"max_steps": 15000,
|
| 251 |
+
"num_input_tokens_seen": 0,
|
| 252 |
+
"num_train_epochs": 3,
|
| 253 |
+
"save_steps": 500,
|
| 254 |
+
"stateful_callbacks": {
|
| 255 |
+
"TrainerControl": {
|
| 256 |
+
"args": {
|
| 257 |
+
"should_epoch_stop": false,
|
| 258 |
+
"should_evaluate": false,
|
| 259 |
+
"should_log": false,
|
| 260 |
+
"should_save": true,
|
| 261 |
+
"should_training_stop": true
|
| 262 |
+
},
|
| 263 |
+
"attributes": {}
|
| 264 |
+
}
|
| 265 |
+
},
|
| 266 |
+
"total_flos": 0.0,
|
| 267 |
+
"train_batch_size": 8,
|
| 268 |
+
"trial_name": null,
|
| 269 |
+
"trial_params": null
|
| 270 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:544708f8849171ce4672556f780bc06631264a1bdc1140d0e6c0702ef87caa13
|
| 3 |
+
size 5432
|