Training in progress, epoch 2
Browse files- all_results.json +5 -5
- model.safetensors +1 -1
- runs/Jan15_20-00-25_a99b99d67614/events.out.tfevents.1736971226.a99b99d67614.505.23 +3 -0
- runs/Jan15_20-03-54_a99b99d67614/events.out.tfevents.1736971435.a99b99d67614.505.24 +3 -0
- runs/Jan15_20-04-21_a99b99d67614/events.out.tfevents.1736971462.a99b99d67614.505.25 +3 -0
- runs/Jan15_20-04-45_a99b99d67614/events.out.tfevents.1736971487.a99b99d67614.505.26 +3 -0
- runs/Jan15_20-06-01_a99b99d67614/events.out.tfevents.1736971563.a99b99d67614.505.27 +3 -0
- runs/Jan15_20-09-35_a99b99d67614/events.out.tfevents.1736971776.a99b99d67614.505.28 +3 -0
- runs/Jan15_20-11-17_a99b99d67614/events.out.tfevents.1736971878.a99b99d67614.505.29 +3 -0
- runs/Jan15_20-12-32_a99b99d67614/events.out.tfevents.1736971953.a99b99d67614.505.30 +3 -0
- runs/Jan15_20-13-32_a99b99d67614/events.out.tfevents.1736972014.a99b99d67614.505.31 +3 -0
- runs/Jan15_20-14-36_a99b99d67614/events.out.tfevents.1736972077.a99b99d67614.505.32 +3 -0
- train_results.json +5 -5
- trainer_state.json +341 -208
- training_args.bin +1 -1
all_results.json
CHANGED
|
@@ -5,9 +5,9 @@
|
|
| 5 |
"eval_runtime": 3.9702,
|
| 6 |
"eval_samples_per_second": 73.548,
|
| 7 |
"eval_steps_per_second": 9.319,
|
| 8 |
-
"total_flos":
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second":
|
| 13 |
}
|
|
|
|
| 5 |
"eval_runtime": 3.9702,
|
| 6 |
"eval_samples_per_second": 73.548,
|
| 7 |
"eval_steps_per_second": 9.319,
|
| 8 |
+
"total_flos": 5.916629591779738e+17,
|
| 9 |
+
"train_loss": 0.020844636825805014,
|
| 10 |
+
"train_runtime": 401.6875,
|
| 11 |
+
"train_samples_per_second": 18.621,
|
| 12 |
+
"train_steps_per_second": 1.17
|
| 13 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 350497036
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:04baf4507a8e1db4fa09c84c382f7ffb623cdc85a6428e220660a802d73b633b
|
| 3 |
size 350497036
|
runs/Jan15_20-00-25_a99b99d67614/events.out.tfevents.1736971226.a99b99d67614.505.23
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4136644bdda2b28ecc7c8056dd4a0c62f1045ced8fd4fc4f446d34b810c915ea
|
| 3 |
+
size 16665
|
runs/Jan15_20-03-54_a99b99d67614/events.out.tfevents.1736971435.a99b99d67614.505.24
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:07e4d97f320983e3d0ffbdc2a4bed50d41e082851563e5df7520a88e89cf137c
|
| 3 |
+
size 16693
|
runs/Jan15_20-04-21_a99b99d67614/events.out.tfevents.1736971462.a99b99d67614.505.25
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:104574891a791fd31d7546229a10d159f6261f828d19f2d9687d194ea2bf6129
|
| 3 |
+
size 16693
|
runs/Jan15_20-04-45_a99b99d67614/events.out.tfevents.1736971487.a99b99d67614.505.26
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f6fa2a8e661341099c7d0c3e9e7ddaff51111c99cced0dedd6ac68ed019b1078
|
| 3 |
+
size 16665
|
runs/Jan15_20-06-01_a99b99d67614/events.out.tfevents.1736971563.a99b99d67614.505.27
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:83f3cf9beb83507f16417ee9009bd49ed7af0b3dd56599fde0a4b45e961f6129
|
| 3 |
+
size 17521
|
runs/Jan15_20-09-35_a99b99d67614/events.out.tfevents.1736971776.a99b99d67614.505.28
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b8b01367ff3620586805a4a4a2292429adf9396baa9a64913c564bf7875218a5
|
| 3 |
+
size 17569
|
runs/Jan15_20-11-17_a99b99d67614/events.out.tfevents.1736971878.a99b99d67614.505.29
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ea36559b72d0b9b6e09066ac12a0b159775173f187db37dba87bfbea5db232e
|
| 3 |
+
size 17569
|
runs/Jan15_20-12-32_a99b99d67614/events.out.tfevents.1736971953.a99b99d67614.505.30
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cc4e6eb67a94ecee1c6c3de970aa120f2264f7be6bf7359830688274bbd8d47d
|
| 3 |
+
size 17569
|
runs/Jan15_20-13-32_a99b99d67614/events.out.tfevents.1736972014.a99b99d67614.505.31
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fe679ed8609dd968498732b3a3b28ab9e6ec8a8eba64769fee4109ccc20b0fd4
|
| 3 |
+
size 17569
|
runs/Jan15_20-14-36_a99b99d67614/events.out.tfevents.1736972077.a99b99d67614.505.32
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:351ebe821b37bf1d768caf8f0c7678bc5d68d54ffe37ec935636c01eec50d914
|
| 3 |
+
size 19238
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"total_flos": 5.916629591779738e+17,
|
| 4 |
+
"train_loss": 0.020844636825805014,
|
| 5 |
+
"train_runtime": 401.6875,
|
| 6 |
+
"train_samples_per_second": 18.621,
|
| 7 |
+
"train_steps_per_second": 1.17
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,311 +1,444 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./logo-matching-base/checkpoint-
|
| 4 |
"epoch": 10.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 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.
|
| 13 |
-
"grad_norm": 0.
|
| 14 |
-
"learning_rate": 0.
|
| 15 |
-
"loss": 0.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm": 0.
|
| 21 |
-
"learning_rate": 0.
|
| 22 |
-
"loss": 0.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch":
|
| 27 |
-
"
|
| 28 |
-
"
|
| 29 |
-
"
|
| 30 |
-
"eval_samples_per_second": 66.184,
|
| 31 |
-
"eval_steps_per_second": 8.411,
|
| 32 |
-
"step": 28
|
| 33 |
-
},
|
| 34 |
-
{
|
| 35 |
-
"epoch": 1.0714285714285714,
|
| 36 |
-
"grad_norm": 0.057985421270132065,
|
| 37 |
-
"learning_rate": 0.0001785714285714286,
|
| 38 |
-
"loss": 0.0461,
|
| 39 |
"step": 30
|
| 40 |
},
|
| 41 |
{
|
| 42 |
-
"epoch":
|
| 43 |
-
"grad_norm": 0.
|
| 44 |
-
"learning_rate": 0.
|
| 45 |
-
"loss": 0.
|
| 46 |
"step": 40
|
| 47 |
},
|
| 48 |
{
|
| 49 |
-
"epoch": 1.
|
| 50 |
-
"
|
| 51 |
-
"
|
| 52 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 53 |
"step": 50
|
| 54 |
},
|
| 55 |
{
|
| 56 |
-
"epoch":
|
| 57 |
-
"
|
| 58 |
-
"
|
| 59 |
-
"
|
| 60 |
-
"eval_samples_per_second": 92.718,
|
| 61 |
-
"eval_steps_per_second": 11.784,
|
| 62 |
-
"step": 56
|
| 63 |
-
},
|
| 64 |
-
{
|
| 65 |
-
"epoch": 2.142857142857143,
|
| 66 |
-
"grad_norm": 0.02560759335756302,
|
| 67 |
-
"learning_rate": 0.00015714285714285716,
|
| 68 |
-
"loss": 0.0174,
|
| 69 |
"step": 60
|
| 70 |
},
|
| 71 |
{
|
| 72 |
-
"epoch":
|
| 73 |
-
"grad_norm": 0.
|
| 74 |
-
"learning_rate": 0.
|
| 75 |
-
"loss": 0.
|
| 76 |
"step": 70
|
| 77 |
},
|
| 78 |
{
|
| 79 |
-
"epoch":
|
| 80 |
-
"grad_norm": 0.
|
| 81 |
-
"learning_rate": 0.
|
| 82 |
-
"loss": 0.
|
| 83 |
"step": 80
|
| 84 |
},
|
| 85 |
{
|
| 86 |
-
"epoch":
|
| 87 |
-
"
|
| 88 |
-
"
|
| 89 |
-
"
|
| 90 |
-
"eval_samples_per_second": 102.924,
|
| 91 |
-
"eval_steps_per_second": 13.081,
|
| 92 |
-
"step": 84
|
| 93 |
-
},
|
| 94 |
-
{
|
| 95 |
-
"epoch": 3.2142857142857144,
|
| 96 |
-
"grad_norm": 0.019254567101597786,
|
| 97 |
-
"learning_rate": 0.00013571428571428572,
|
| 98 |
-
"loss": 0.0145,
|
| 99 |
"step": 90
|
| 100 |
},
|
| 101 |
{
|
| 102 |
-
"epoch":
|
| 103 |
-
"
|
| 104 |
-
"
|
| 105 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 106 |
"step": 100
|
| 107 |
},
|
| 108 |
{
|
| 109 |
-
"epoch":
|
| 110 |
-
"grad_norm": 0.
|
| 111 |
-
"learning_rate": 0.
|
| 112 |
-
"loss": 0.
|
| 113 |
"step": 110
|
| 114 |
},
|
| 115 |
{
|
| 116 |
-
"epoch":
|
| 117 |
-
"
|
| 118 |
-
"
|
| 119 |
-
"
|
| 120 |
-
"eval_samples_per_second": 66.797,
|
| 121 |
-
"eval_steps_per_second": 8.489,
|
| 122 |
-
"step": 112
|
| 123 |
-
},
|
| 124 |
-
{
|
| 125 |
-
"epoch": 4.285714285714286,
|
| 126 |
-
"grad_norm": 0.02770456299185753,
|
| 127 |
-
"learning_rate": 0.00011428571428571428,
|
| 128 |
-
"loss": 0.0145,
|
| 129 |
"step": 120
|
| 130 |
},
|
| 131 |
{
|
| 132 |
-
"epoch":
|
| 133 |
-
"grad_norm": 0.
|
| 134 |
-
"learning_rate": 0.
|
| 135 |
-
"loss": 0.
|
| 136 |
"step": 130
|
| 137 |
},
|
| 138 |
{
|
| 139 |
-
"epoch":
|
| 140 |
-
"grad_norm": 0.
|
| 141 |
-
"learning_rate": 0.
|
| 142 |
-
"loss": 0.
|
| 143 |
"step": 140
|
| 144 |
},
|
| 145 |
{
|
| 146 |
-
"epoch":
|
| 147 |
-
"eval_accuracy": 0.
|
| 148 |
-
"eval_loss": 0.
|
| 149 |
-
"eval_runtime":
|
| 150 |
-
"eval_samples_per_second":
|
| 151 |
-
"eval_steps_per_second": 11.
|
| 152 |
-
"step":
|
| 153 |
-
},
|
| 154 |
-
{
|
| 155 |
-
"epoch":
|
| 156 |
-
"grad_norm": 0.
|
| 157 |
-
"learning_rate":
|
| 158 |
-
"loss": 0.
|
| 159 |
"step": 150
|
| 160 |
},
|
| 161 |
{
|
| 162 |
-
"epoch":
|
| 163 |
-
"grad_norm": 0.
|
| 164 |
-
"learning_rate":
|
| 165 |
-
"loss": 0.
|
| 166 |
"step": 160
|
| 167 |
},
|
| 168 |
{
|
| 169 |
-
"epoch":
|
| 170 |
-
"
|
| 171 |
-
"
|
| 172 |
-
"
|
| 173 |
-
"eval_samples_per_second": 108.154,
|
| 174 |
-
"eval_steps_per_second": 13.745,
|
| 175 |
-
"step": 168
|
| 176 |
-
},
|
| 177 |
-
{
|
| 178 |
-
"epoch": 6.071428571428571,
|
| 179 |
-
"grad_norm": 0.026734082028269768,
|
| 180 |
-
"learning_rate": 7.857142857142858e-05,
|
| 181 |
-
"loss": 0.0129,
|
| 182 |
"step": 170
|
| 183 |
},
|
| 184 |
{
|
| 185 |
-
"epoch":
|
| 186 |
-
"grad_norm": 0.
|
| 187 |
-
"learning_rate":
|
| 188 |
-
"loss": 0.
|
| 189 |
"step": 180
|
| 190 |
},
|
| 191 |
{
|
| 192 |
-
"epoch":
|
| 193 |
-
"
|
| 194 |
-
"
|
| 195 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 196 |
"step": 190
|
| 197 |
},
|
| 198 |
{
|
| 199 |
-
"epoch":
|
| 200 |
-
"
|
| 201 |
-
"
|
| 202 |
-
"
|
| 203 |
-
"eval_samples_per_second": 80.757,
|
| 204 |
-
"eval_steps_per_second": 10.263,
|
| 205 |
-
"step": 196
|
| 206 |
-
},
|
| 207 |
-
{
|
| 208 |
-
"epoch": 7.142857142857143,
|
| 209 |
-
"grad_norm": 0.023527879267930984,
|
| 210 |
-
"learning_rate": 5.714285714285714e-05,
|
| 211 |
-
"loss": 0.0116,
|
| 212 |
"step": 200
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"epoch":
|
| 216 |
-
"grad_norm": 0.
|
| 217 |
-
"learning_rate":
|
| 218 |
-
"loss": 0.
|
| 219 |
"step": 210
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch":
|
| 223 |
-
"grad_norm": 0.
|
| 224 |
-
"learning_rate":
|
| 225 |
-
"loss": 0.
|
| 226 |
"step": 220
|
| 227 |
},
|
| 228 |
{
|
| 229 |
-
"epoch":
|
| 230 |
-
"
|
| 231 |
-
"
|
| 232 |
-
"
|
| 233 |
-
"eval_samples_per_second": 106.173,
|
| 234 |
-
"eval_steps_per_second": 13.494,
|
| 235 |
-
"step": 224
|
| 236 |
-
},
|
| 237 |
-
{
|
| 238 |
-
"epoch": 8.214285714285714,
|
| 239 |
-
"grad_norm": 0.023050658404827118,
|
| 240 |
-
"learning_rate": 3.571428571428572e-05,
|
| 241 |
-
"loss": 0.0124,
|
| 242 |
"step": 230
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch":
|
| 246 |
-
"
|
| 247 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 248 |
"loss": 0.0118,
|
| 249 |
"step": 240
|
| 250 |
},
|
| 251 |
{
|
| 252 |
-
"epoch":
|
| 253 |
-
"grad_norm": 0.
|
| 254 |
-
"learning_rate":
|
| 255 |
-
"loss": 0.
|
| 256 |
"step": 250
|
| 257 |
},
|
| 258 |
{
|
| 259 |
-
"epoch":
|
| 260 |
-
"
|
| 261 |
-
"
|
| 262 |
-
"
|
| 263 |
-
"eval_samples_per_second": 105.789,
|
| 264 |
-
"eval_steps_per_second": 13.445,
|
| 265 |
-
"step": 252
|
| 266 |
-
},
|
| 267 |
-
{
|
| 268 |
-
"epoch": 9.285714285714286,
|
| 269 |
-
"grad_norm": 0.02203691191971302,
|
| 270 |
-
"learning_rate": 1.4285714285714285e-05,
|
| 271 |
-
"loss": 0.0126,
|
| 272 |
"step": 260
|
| 273 |
},
|
| 274 |
{
|
| 275 |
-
"epoch":
|
| 276 |
-
"grad_norm": 0.
|
| 277 |
-
"learning_rate":
|
| 278 |
-
"loss": 0.
|
| 279 |
"step": 270
|
| 280 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 281 |
{
|
| 282 |
"epoch": 10.0,
|
| 283 |
-
"grad_norm": 0.
|
| 284 |
"learning_rate": 0.0,
|
| 285 |
-
"loss": 0.
|
| 286 |
-
"step":
|
| 287 |
},
|
| 288 |
{
|
| 289 |
"epoch": 10.0,
|
| 290 |
-
"eval_accuracy": 0.
|
| 291 |
-
"eval_loss": 0.
|
| 292 |
-
"eval_runtime":
|
| 293 |
-
"eval_samples_per_second":
|
| 294 |
-
"eval_steps_per_second":
|
| 295 |
-
"step":
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 10.0,
|
| 299 |
-
"step":
|
| 300 |
-
"total_flos":
|
| 301 |
-
"train_loss": 0.
|
| 302 |
-
"train_runtime":
|
| 303 |
-
"train_samples_per_second":
|
| 304 |
-
"train_steps_per_second":
|
| 305 |
}
|
| 306 |
],
|
| 307 |
"logging_steps": 10,
|
| 308 |
-
"max_steps":
|
| 309 |
"num_input_tokens_seen": 0,
|
| 310 |
"num_train_epochs": 10,
|
| 311 |
"save_steps": 500,
|
|
@@ -321,7 +454,7 @@
|
|
| 321 |
"attributes": {}
|
| 322 |
}
|
| 323 |
},
|
| 324 |
-
"total_flos":
|
| 325 |
"train_batch_size": 16,
|
| 326 |
"trial_name": null,
|
| 327 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.03154107183218002,
|
| 3 |
+
"best_model_checkpoint": "./logo-matching-base/checkpoint-141",
|
| 4 |
"epoch": 10.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 470,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.2127659574468085,
|
| 13 |
+
"grad_norm": 0.20348502695560455,
|
| 14 |
+
"learning_rate": 0.00019574468085106384,
|
| 15 |
+
"loss": 0.2136,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.425531914893617,
|
| 20 |
+
"grad_norm": 0.20952267944812775,
|
| 21 |
+
"learning_rate": 0.00019148936170212768,
|
| 22 |
+
"loss": 0.0284,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.6382978723404256,
|
| 27 |
+
"grad_norm": 0.21442832052707672,
|
| 28 |
+
"learning_rate": 0.0001872340425531915,
|
| 29 |
+
"loss": 0.183,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.851063829787234,
|
| 34 |
+
"grad_norm": 0.38694852590560913,
|
| 35 |
+
"learning_rate": 0.00018297872340425532,
|
| 36 |
+
"loss": 0.0225,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 1.0,
|
| 41 |
+
"eval_accuracy": 0.3229357798165138,
|
| 42 |
+
"eval_loss": 0.032035112380981445,
|
| 43 |
+
"eval_runtime": 6.0654,
|
| 44 |
+
"eval_samples_per_second": 89.854,
|
| 45 |
+
"eval_steps_per_second": 11.376,
|
| 46 |
+
"step": 47
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.0638297872340425,
|
| 50 |
+
"grad_norm": 0.11812812834978104,
|
| 51 |
+
"learning_rate": 0.00017872340425531915,
|
| 52 |
+
"loss": 0.02,
|
| 53 |
"step": 50
|
| 54 |
},
|
| 55 |
{
|
| 56 |
+
"epoch": 1.2765957446808511,
|
| 57 |
+
"grad_norm": 0.1575324535369873,
|
| 58 |
+
"learning_rate": 0.00017446808510638298,
|
| 59 |
+
"loss": 0.0213,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 60 |
"step": 60
|
| 61 |
},
|
| 62 |
{
|
| 63 |
+
"epoch": 1.4893617021276595,
|
| 64 |
+
"grad_norm": 0.12748625874519348,
|
| 65 |
+
"learning_rate": 0.00017021276595744682,
|
| 66 |
+
"loss": 0.0207,
|
| 67 |
"step": 70
|
| 68 |
},
|
| 69 |
{
|
| 70 |
+
"epoch": 1.702127659574468,
|
| 71 |
+
"grad_norm": 0.11703016608953476,
|
| 72 |
+
"learning_rate": 0.00016595744680851065,
|
| 73 |
+
"loss": 0.0178,
|
| 74 |
"step": 80
|
| 75 |
},
|
| 76 |
{
|
| 77 |
+
"epoch": 1.9148936170212765,
|
| 78 |
+
"grad_norm": 0.09279447048902512,
|
| 79 |
+
"learning_rate": 0.00016170212765957446,
|
| 80 |
+
"loss": 0.0184,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 81 |
"step": 90
|
| 82 |
},
|
| 83 |
{
|
| 84 |
+
"epoch": 2.0,
|
| 85 |
+
"eval_accuracy": 0.27706422018348625,
|
| 86 |
+
"eval_loss": 0.03198189660906792,
|
| 87 |
+
"eval_runtime": 4.7267,
|
| 88 |
+
"eval_samples_per_second": 115.302,
|
| 89 |
+
"eval_steps_per_second": 14.598,
|
| 90 |
+
"step": 94
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 2.127659574468085,
|
| 94 |
+
"grad_norm": 0.09347425401210785,
|
| 95 |
+
"learning_rate": 0.00015744680851063832,
|
| 96 |
+
"loss": 0.0161,
|
| 97 |
"step": 100
|
| 98 |
},
|
| 99 |
{
|
| 100 |
+
"epoch": 2.3404255319148937,
|
| 101 |
+
"grad_norm": 0.09166835993528366,
|
| 102 |
+
"learning_rate": 0.00015319148936170213,
|
| 103 |
+
"loss": 0.0169,
|
| 104 |
"step": 110
|
| 105 |
},
|
| 106 |
{
|
| 107 |
+
"epoch": 2.5531914893617023,
|
| 108 |
+
"grad_norm": 0.12563753128051758,
|
| 109 |
+
"learning_rate": 0.00014893617021276596,
|
| 110 |
+
"loss": 0.0181,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 111 |
"step": 120
|
| 112 |
},
|
| 113 |
{
|
| 114 |
+
"epoch": 2.7659574468085104,
|
| 115 |
+
"grad_norm": 0.10614734888076782,
|
| 116 |
+
"learning_rate": 0.0001446808510638298,
|
| 117 |
+
"loss": 0.0166,
|
| 118 |
"step": 130
|
| 119 |
},
|
| 120 |
{
|
| 121 |
+
"epoch": 2.978723404255319,
|
| 122 |
+
"grad_norm": 0.10630539804697037,
|
| 123 |
+
"learning_rate": 0.00014042553191489363,
|
| 124 |
+
"loss": 0.0177,
|
| 125 |
"step": 140
|
| 126 |
},
|
| 127 |
{
|
| 128 |
+
"epoch": 3.0,
|
| 129 |
+
"eval_accuracy": 0.26788990825688075,
|
| 130 |
+
"eval_loss": 0.03154107183218002,
|
| 131 |
+
"eval_runtime": 6.1724,
|
| 132 |
+
"eval_samples_per_second": 88.296,
|
| 133 |
+
"eval_steps_per_second": 11.179,
|
| 134 |
+
"step": 141
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.1914893617021276,
|
| 138 |
+
"grad_norm": 0.08008704334497452,
|
| 139 |
+
"learning_rate": 0.00013617021276595746,
|
| 140 |
+
"loss": 0.0175,
|
| 141 |
"step": 150
|
| 142 |
},
|
| 143 |
{
|
| 144 |
+
"epoch": 3.404255319148936,
|
| 145 |
+
"grad_norm": 0.1296830177307129,
|
| 146 |
+
"learning_rate": 0.00013191489361702127,
|
| 147 |
+
"loss": 0.0174,
|
| 148 |
"step": 160
|
| 149 |
},
|
| 150 |
{
|
| 151 |
+
"epoch": 3.617021276595745,
|
| 152 |
+
"grad_norm": 0.11523136496543884,
|
| 153 |
+
"learning_rate": 0.00012765957446808513,
|
| 154 |
+
"loss": 0.014,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 155 |
"step": 170
|
| 156 |
},
|
| 157 |
{
|
| 158 |
+
"epoch": 3.829787234042553,
|
| 159 |
+
"grad_norm": 0.10043615847826004,
|
| 160 |
+
"learning_rate": 0.00012340425531914893,
|
| 161 |
+
"loss": 0.0144,
|
| 162 |
"step": 180
|
| 163 |
},
|
| 164 |
{
|
| 165 |
+
"epoch": 4.0,
|
| 166 |
+
"eval_accuracy": 0.25321100917431194,
|
| 167 |
+
"eval_loss": 0.03288332372903824,
|
| 168 |
+
"eval_runtime": 5.7194,
|
| 169 |
+
"eval_samples_per_second": 95.289,
|
| 170 |
+
"eval_steps_per_second": 12.064,
|
| 171 |
+
"step": 188
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 4.042553191489362,
|
| 175 |
+
"grad_norm": 0.1341152787208557,
|
| 176 |
+
"learning_rate": 0.00011914893617021277,
|
| 177 |
+
"loss": 0.0143,
|
| 178 |
"step": 190
|
| 179 |
},
|
| 180 |
{
|
| 181 |
+
"epoch": 4.25531914893617,
|
| 182 |
+
"grad_norm": 0.11163297295570374,
|
| 183 |
+
"learning_rate": 0.00011489361702127661,
|
| 184 |
+
"loss": 0.0147,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 185 |
"step": 200
|
| 186 |
},
|
| 187 |
{
|
| 188 |
+
"epoch": 4.468085106382979,
|
| 189 |
+
"grad_norm": 0.14031550288200378,
|
| 190 |
+
"learning_rate": 0.00011063829787234043,
|
| 191 |
+
"loss": 0.0129,
|
| 192 |
"step": 210
|
| 193 |
},
|
| 194 |
{
|
| 195 |
+
"epoch": 4.680851063829787,
|
| 196 |
+
"grad_norm": 0.11095025390386581,
|
| 197 |
+
"learning_rate": 0.00010638297872340425,
|
| 198 |
+
"loss": 0.0159,
|
| 199 |
"step": 220
|
| 200 |
},
|
| 201 |
{
|
| 202 |
+
"epoch": 4.8936170212765955,
|
| 203 |
+
"grad_norm": 0.09944932162761688,
|
| 204 |
+
"learning_rate": 0.00010212765957446809,
|
| 205 |
+
"loss": 0.0133,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 206 |
"step": 230
|
| 207 |
},
|
| 208 |
{
|
| 209 |
+
"epoch": 5.0,
|
| 210 |
+
"eval_accuracy": 0.30825688073394497,
|
| 211 |
+
"eval_loss": 0.03211478143930435,
|
| 212 |
+
"eval_runtime": 4.8718,
|
| 213 |
+
"eval_samples_per_second": 111.869,
|
| 214 |
+
"eval_steps_per_second": 14.163,
|
| 215 |
+
"step": 235
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 5.1063829787234045,
|
| 219 |
+
"grad_norm": 0.07057506591081619,
|
| 220 |
+
"learning_rate": 9.787234042553192e-05,
|
| 221 |
"loss": 0.0118,
|
| 222 |
"step": 240
|
| 223 |
},
|
| 224 |
{
|
| 225 |
+
"epoch": 5.319148936170213,
|
| 226 |
+
"grad_norm": 0.11148671805858612,
|
| 227 |
+
"learning_rate": 9.361702127659576e-05,
|
| 228 |
+
"loss": 0.0123,
|
| 229 |
"step": 250
|
| 230 |
},
|
| 231 |
{
|
| 232 |
+
"epoch": 5.531914893617021,
|
| 233 |
+
"grad_norm": 0.1193113625049591,
|
| 234 |
+
"learning_rate": 8.936170212765958e-05,
|
| 235 |
+
"loss": 0.0128,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 236 |
"step": 260
|
| 237 |
},
|
| 238 |
{
|
| 239 |
+
"epoch": 5.74468085106383,
|
| 240 |
+
"grad_norm": 0.15620863437652588,
|
| 241 |
+
"learning_rate": 8.510638297872341e-05,
|
| 242 |
+
"loss": 0.013,
|
| 243 |
"step": 270
|
| 244 |
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 5.957446808510638,
|
| 247 |
+
"grad_norm": 0.09126376360654831,
|
| 248 |
+
"learning_rate": 8.085106382978723e-05,
|
| 249 |
+
"loss": 0.0108,
|
| 250 |
+
"step": 280
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 6.0,
|
| 254 |
+
"eval_accuracy": 0.3137614678899083,
|
| 255 |
+
"eval_loss": 0.03295579180121422,
|
| 256 |
+
"eval_runtime": 4.8988,
|
| 257 |
+
"eval_samples_per_second": 111.251,
|
| 258 |
+
"eval_steps_per_second": 14.085,
|
| 259 |
+
"step": 282
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 6.170212765957447,
|
| 263 |
+
"grad_norm": 0.11392467468976974,
|
| 264 |
+
"learning_rate": 7.659574468085106e-05,
|
| 265 |
+
"loss": 0.0115,
|
| 266 |
+
"step": 290
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 6.382978723404255,
|
| 270 |
+
"grad_norm": 0.07867500931024551,
|
| 271 |
+
"learning_rate": 7.23404255319149e-05,
|
| 272 |
+
"loss": 0.0105,
|
| 273 |
+
"step": 300
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 6.595744680851064,
|
| 277 |
+
"grad_norm": 0.07005509734153748,
|
| 278 |
+
"learning_rate": 6.808510638297873e-05,
|
| 279 |
+
"loss": 0.0099,
|
| 280 |
+
"step": 310
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 6.808510638297872,
|
| 284 |
+
"grad_norm": 0.10206523537635803,
|
| 285 |
+
"learning_rate": 6.382978723404256e-05,
|
| 286 |
+
"loss": 0.0097,
|
| 287 |
+
"step": 320
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 7.0,
|
| 291 |
+
"eval_accuracy": 0.25688073394495414,
|
| 292 |
+
"eval_loss": 0.034237515181303024,
|
| 293 |
+
"eval_runtime": 5.8682,
|
| 294 |
+
"eval_samples_per_second": 92.873,
|
| 295 |
+
"eval_steps_per_second": 11.758,
|
| 296 |
+
"step": 329
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 7.0212765957446805,
|
| 300 |
+
"grad_norm": 0.15130436420440674,
|
| 301 |
+
"learning_rate": 5.9574468085106384e-05,
|
| 302 |
+
"loss": 0.0106,
|
| 303 |
+
"step": 330
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 7.23404255319149,
|
| 307 |
+
"grad_norm": 0.07350896298885345,
|
| 308 |
+
"learning_rate": 5.531914893617022e-05,
|
| 309 |
+
"loss": 0.0086,
|
| 310 |
+
"step": 340
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 7.446808510638298,
|
| 314 |
+
"grad_norm": 0.10417384654283524,
|
| 315 |
+
"learning_rate": 5.1063829787234044e-05,
|
| 316 |
+
"loss": 0.0108,
|
| 317 |
+
"step": 350
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 7.659574468085106,
|
| 321 |
+
"grad_norm": 0.12092699855566025,
|
| 322 |
+
"learning_rate": 4.680851063829788e-05,
|
| 323 |
+
"loss": 0.0079,
|
| 324 |
+
"step": 360
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 7.872340425531915,
|
| 328 |
+
"grad_norm": 0.11693856120109558,
|
| 329 |
+
"learning_rate": 4.2553191489361704e-05,
|
| 330 |
+
"loss": 0.0086,
|
| 331 |
+
"step": 370
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 8.0,
|
| 335 |
+
"eval_accuracy": 0.3192660550458716,
|
| 336 |
+
"eval_loss": 0.03412623330950737,
|
| 337 |
+
"eval_runtime": 6.8764,
|
| 338 |
+
"eval_samples_per_second": 79.257,
|
| 339 |
+
"eval_steps_per_second": 10.034,
|
| 340 |
+
"step": 376
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 8.085106382978724,
|
| 344 |
+
"grad_norm": 0.07324172556400299,
|
| 345 |
+
"learning_rate": 3.829787234042553e-05,
|
| 346 |
+
"loss": 0.0076,
|
| 347 |
+
"step": 380
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 8.297872340425531,
|
| 351 |
+
"grad_norm": 0.1128627359867096,
|
| 352 |
+
"learning_rate": 3.4042553191489365e-05,
|
| 353 |
+
"loss": 0.0074,
|
| 354 |
+
"step": 390
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 8.51063829787234,
|
| 358 |
+
"grad_norm": 0.10660151392221451,
|
| 359 |
+
"learning_rate": 2.9787234042553192e-05,
|
| 360 |
+
"loss": 0.0067,
|
| 361 |
+
"step": 400
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 8.72340425531915,
|
| 365 |
+
"grad_norm": 0.0888807401061058,
|
| 366 |
+
"learning_rate": 2.5531914893617022e-05,
|
| 367 |
+
"loss": 0.0076,
|
| 368 |
+
"step": 410
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 8.936170212765958,
|
| 372 |
+
"grad_norm": 0.07239257544279099,
|
| 373 |
+
"learning_rate": 2.1276595744680852e-05,
|
| 374 |
+
"loss": 0.0068,
|
| 375 |
+
"step": 420
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 9.0,
|
| 379 |
+
"eval_accuracy": 0.3155963302752294,
|
| 380 |
+
"eval_loss": 0.03542930632829666,
|
| 381 |
+
"eval_runtime": 5.9234,
|
| 382 |
+
"eval_samples_per_second": 92.008,
|
| 383 |
+
"eval_steps_per_second": 11.649,
|
| 384 |
+
"step": 423
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 9.148936170212766,
|
| 388 |
+
"grad_norm": 0.08497753739356995,
|
| 389 |
+
"learning_rate": 1.7021276595744682e-05,
|
| 390 |
+
"loss": 0.0067,
|
| 391 |
+
"step": 430
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 9.361702127659575,
|
| 395 |
+
"grad_norm": 0.06389721482992172,
|
| 396 |
+
"learning_rate": 1.2765957446808511e-05,
|
| 397 |
+
"loss": 0.0062,
|
| 398 |
+
"step": 440
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 9.574468085106384,
|
| 402 |
+
"grad_norm": 0.06799926608800888,
|
| 403 |
+
"learning_rate": 8.510638297872341e-06,
|
| 404 |
+
"loss": 0.0054,
|
| 405 |
+
"step": 450
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 9.787234042553191,
|
| 409 |
+
"grad_norm": 0.08153792470693588,
|
| 410 |
+
"learning_rate": 4.255319148936171e-06,
|
| 411 |
+
"loss": 0.0053,
|
| 412 |
+
"step": 460
|
| 413 |
+
},
|
| 414 |
{
|
| 415 |
"epoch": 10.0,
|
| 416 |
+
"grad_norm": 0.1752246469259262,
|
| 417 |
"learning_rate": 0.0,
|
| 418 |
+
"loss": 0.0056,
|
| 419 |
+
"step": 470
|
| 420 |
},
|
| 421 |
{
|
| 422 |
"epoch": 10.0,
|
| 423 |
+
"eval_accuracy": 0.3339449541284404,
|
| 424 |
+
"eval_loss": 0.03554193675518036,
|
| 425 |
+
"eval_runtime": 6.6038,
|
| 426 |
+
"eval_samples_per_second": 82.529,
|
| 427 |
+
"eval_steps_per_second": 10.449,
|
| 428 |
+
"step": 470
|
| 429 |
},
|
| 430 |
{
|
| 431 |
"epoch": 10.0,
|
| 432 |
+
"step": 470,
|
| 433 |
+
"total_flos": 5.916629591779738e+17,
|
| 434 |
+
"train_loss": 0.020844636825805014,
|
| 435 |
+
"train_runtime": 401.6875,
|
| 436 |
+
"train_samples_per_second": 18.621,
|
| 437 |
+
"train_steps_per_second": 1.17
|
| 438 |
}
|
| 439 |
],
|
| 440 |
"logging_steps": 10,
|
| 441 |
+
"max_steps": 470,
|
| 442 |
"num_input_tokens_seen": 0,
|
| 443 |
"num_train_epochs": 10,
|
| 444 |
"save_steps": 500,
|
|
|
|
| 454 |
"attributes": {}
|
| 455 |
}
|
| 456 |
},
|
| 457 |
+
"total_flos": 5.916629591779738e+17,
|
| 458 |
"train_batch_size": 16,
|
| 459 |
"trial_name": null,
|
| 460 |
"trial_params": null
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5368
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a54f12faa92c2188517ce8459d7a38b184f5075cca10aea02a378d12b2444aa4
|
| 3 |
size 5368
|