Upload 2 files
Browse files
MoA-150M_results.json
ADDED
|
@@ -0,0 +1,331 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"entropy": 2.184988856315613,
|
| 4 |
+
"epoch": 0.0625,
|
| 5 |
+
"grad_norm": 6.174712657928467,
|
| 6 |
+
"learning_rate": 0.00029531249999999995,
|
| 7 |
+
"loss": 0.513,
|
| 8 |
+
"mean_token_accuracy": 0.9921568632125854,
|
| 9 |
+
"num_tokens": 4096.0,
|
| 10 |
+
"step": 2
|
| 11 |
+
},
|
| 12 |
+
{
|
| 13 |
+
"entropy": 1.7302125096321106,
|
| 14 |
+
"epoch": 0.125,
|
| 15 |
+
"grad_norm": 0.9681358933448792,
|
| 16 |
+
"learning_rate": 0.00028593749999999995,
|
| 17 |
+
"loss": 0.441,
|
| 18 |
+
"mean_token_accuracy": 0.9870097935199738,
|
| 19 |
+
"num_tokens": 8192.0,
|
| 20 |
+
"step": 4
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"entropy": 2.0538085103034973,
|
| 24 |
+
"epoch": 0.1875,
|
| 25 |
+
"grad_norm": 0.5153102278709412,
|
| 26 |
+
"learning_rate": 0.00027656249999999995,
|
| 27 |
+
"loss": 0.4929,
|
| 28 |
+
"mean_token_accuracy": 0.9980392158031464,
|
| 29 |
+
"num_tokens": 12288.0,
|
| 30 |
+
"step": 6
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"entropy": 1.7263505458831787,
|
| 34 |
+
"epoch": 0.25,
|
| 35 |
+
"grad_norm": 0.6404310464859009,
|
| 36 |
+
"learning_rate": 0.00026718749999999996,
|
| 37 |
+
"loss": 0.4014,
|
| 38 |
+
"mean_token_accuracy": 0.9987744987010956,
|
| 39 |
+
"num_tokens": 16384.0,
|
| 40 |
+
"step": 8
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"entropy": 1.8347786664962769,
|
| 44 |
+
"epoch": 0.3125,
|
| 45 |
+
"grad_norm": 0.6209350228309631,
|
| 46 |
+
"learning_rate": 0.00025781249999999996,
|
| 47 |
+
"loss": 0.4365,
|
| 48 |
+
"mean_token_accuracy": 1.0,
|
| 49 |
+
"num_tokens": 20480.0,
|
| 50 |
+
"step": 10
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"entropy": 1.4254534244537354,
|
| 54 |
+
"epoch": 0.375,
|
| 55 |
+
"grad_norm": 0.44427844882011414,
|
| 56 |
+
"learning_rate": 0.00024843749999999996,
|
| 57 |
+
"loss": 0.2791,
|
| 58 |
+
"mean_token_accuracy": 1.0,
|
| 59 |
+
"num_tokens": 24576.0,
|
| 60 |
+
"step": 12
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"entropy": 1.8001930117607117,
|
| 64 |
+
"epoch": 0.4375,
|
| 65 |
+
"grad_norm": 0.3619579076766968,
|
| 66 |
+
"learning_rate": 0.0002390625,
|
| 67 |
+
"loss": 0.3962,
|
| 68 |
+
"mean_token_accuracy": 0.9987744987010956,
|
| 69 |
+
"num_tokens": 28672.0,
|
| 70 |
+
"step": 14
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"entropy": 1.6178001761436462,
|
| 74 |
+
"epoch": 0.5,
|
| 75 |
+
"grad_norm": 0.38358834385871887,
|
| 76 |
+
"learning_rate": 0.0002296875,
|
| 77 |
+
"loss": 0.3492,
|
| 78 |
+
"mean_token_accuracy": 1.0,
|
| 79 |
+
"num_tokens": 32768.0,
|
| 80 |
+
"step": 16
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"entropy": 1.661442220211029,
|
| 84 |
+
"epoch": 0.5625,
|
| 85 |
+
"grad_norm": 0.3749903440475464,
|
| 86 |
+
"learning_rate": 0.00022031249999999997,
|
| 87 |
+
"loss": 0.3582,
|
| 88 |
+
"mean_token_accuracy": 1.0,
|
| 89 |
+
"num_tokens": 36864.0,
|
| 90 |
+
"step": 18
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"entropy": 1.5717861652374268,
|
| 94 |
+
"epoch": 0.625,
|
| 95 |
+
"grad_norm": 0.36388659477233887,
|
| 96 |
+
"learning_rate": 0.00021093749999999997,
|
| 97 |
+
"loss": 0.3126,
|
| 98 |
+
"mean_token_accuracy": 1.0,
|
| 99 |
+
"num_tokens": 40960.0,
|
| 100 |
+
"step": 20
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"entropy": 1.5534449219703674,
|
| 104 |
+
"epoch": 0.6875,
|
| 105 |
+
"grad_norm": 0.40969353914260864,
|
| 106 |
+
"learning_rate": 0.00020156249999999997,
|
| 107 |
+
"loss": 0.3478,
|
| 108 |
+
"mean_token_accuracy": 0.9997549057006836,
|
| 109 |
+
"num_tokens": 45056.0,
|
| 110 |
+
"step": 22
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"entropy": 1.5265448689460754,
|
| 114 |
+
"epoch": 0.75,
|
| 115 |
+
"grad_norm": 0.41839736700057983,
|
| 116 |
+
"learning_rate": 0.00019218749999999998,
|
| 117 |
+
"loss": 0.3601,
|
| 118 |
+
"mean_token_accuracy": 0.9995098114013672,
|
| 119 |
+
"num_tokens": 49152.0,
|
| 120 |
+
"step": 24
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"entropy": 2.074858069419861,
|
| 124 |
+
"epoch": 0.8125,
|
| 125 |
+
"grad_norm": 0.44888272881507874,
|
| 126 |
+
"learning_rate": 0.00018281249999999998,
|
| 127 |
+
"loss": 0.4777,
|
| 128 |
+
"mean_token_accuracy": 0.9997549057006836,
|
| 129 |
+
"num_tokens": 53248.0,
|
| 130 |
+
"step": 26
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"entropy": 1.6095194816589355,
|
| 134 |
+
"epoch": 0.875,
|
| 135 |
+
"grad_norm": 0.5080280900001526,
|
| 136 |
+
"learning_rate": 0.00017343749999999998,
|
| 137 |
+
"loss": 0.3859,
|
| 138 |
+
"mean_token_accuracy": 0.9987744987010956,
|
| 139 |
+
"num_tokens": 57344.0,
|
| 140 |
+
"step": 28
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"entropy": 1.7032344341278076,
|
| 144 |
+
"epoch": 0.9375,
|
| 145 |
+
"grad_norm": 0.3624984622001648,
|
| 146 |
+
"learning_rate": 0.00016406249999999998,
|
| 147 |
+
"loss": 0.3576,
|
| 148 |
+
"mean_token_accuracy": 1.0,
|
| 149 |
+
"num_tokens": 61440.0,
|
| 150 |
+
"step": 30
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"entropy": 1.8873920440673828,
|
| 154 |
+
"epoch": 1.0,
|
| 155 |
+
"grad_norm": 0.618506133556366,
|
| 156 |
+
"learning_rate": 0.00015468749999999999,
|
| 157 |
+
"loss": 0.446,
|
| 158 |
+
"mean_token_accuracy": 0.9997549057006836,
|
| 159 |
+
"num_tokens": 64256.0,
|
| 160 |
+
"step": 32
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"entropy": 1.4586840271949768,
|
| 164 |
+
"epoch": 1.0625,
|
| 165 |
+
"grad_norm": 0.3723963797092438,
|
| 166 |
+
"learning_rate": 0.0001453125,
|
| 167 |
+
"loss": 0.2824,
|
| 168 |
+
"mean_token_accuracy": 1.0,
|
| 169 |
+
"num_tokens": 68352.0,
|
| 170 |
+
"step": 34
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"entropy": 1.2832568883895874,
|
| 174 |
+
"epoch": 1.125,
|
| 175 |
+
"grad_norm": 0.3108985424041748,
|
| 176 |
+
"learning_rate": 0.0001359375,
|
| 177 |
+
"loss": 0.2191,
|
| 178 |
+
"mean_token_accuracy": 1.0,
|
| 179 |
+
"num_tokens": 72448.0,
|
| 180 |
+
"step": 36
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"entropy": 1.523368000984192,
|
| 184 |
+
"epoch": 1.1875,
|
| 185 |
+
"grad_norm": 0.3509906232357025,
|
| 186 |
+
"learning_rate": 0.0001265625,
|
| 187 |
+
"loss": 0.3042,
|
| 188 |
+
"mean_token_accuracy": 0.9995098114013672,
|
| 189 |
+
"num_tokens": 76544.0,
|
| 190 |
+
"step": 38
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"entropy": 1.2994396686553955,
|
| 194 |
+
"epoch": 1.25,
|
| 195 |
+
"grad_norm": 0.3014850616455078,
|
| 196 |
+
"learning_rate": 0.0001171875,
|
| 197 |
+
"loss": 0.2456,
|
| 198 |
+
"mean_token_accuracy": 1.0,
|
| 199 |
+
"num_tokens": 80640.0,
|
| 200 |
+
"step": 40
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"entropy": 1.3811439871788025,
|
| 204 |
+
"epoch": 1.3125,
|
| 205 |
+
"grad_norm": 0.32755109667778015,
|
| 206 |
+
"learning_rate": 0.00010781249999999998,
|
| 207 |
+
"loss": 0.2521,
|
| 208 |
+
"mean_token_accuracy": 1.0,
|
| 209 |
+
"num_tokens": 84736.0,
|
| 210 |
+
"step": 42
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"entropy": 1.1994215250015259,
|
| 214 |
+
"epoch": 1.375,
|
| 215 |
+
"grad_norm": 0.24541084468364716,
|
| 216 |
+
"learning_rate": 9.843749999999999e-05,
|
| 217 |
+
"loss": 0.2118,
|
| 218 |
+
"mean_token_accuracy": 1.0,
|
| 219 |
+
"num_tokens": 88832.0,
|
| 220 |
+
"step": 44
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"entropy": 1.0519097447395325,
|
| 224 |
+
"epoch": 1.4375,
|
| 225 |
+
"grad_norm": 0.2063349187374115,
|
| 226 |
+
"learning_rate": 8.906249999999999e-05,
|
| 227 |
+
"loss": 0.1943,
|
| 228 |
+
"mean_token_accuracy": 1.0,
|
| 229 |
+
"num_tokens": 92928.0,
|
| 230 |
+
"step": 46
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"entropy": 1.1152112483978271,
|
| 234 |
+
"epoch": 1.5,
|
| 235 |
+
"grad_norm": 0.31837204098701477,
|
| 236 |
+
"learning_rate": 7.968749999999999e-05,
|
| 237 |
+
"loss": 0.2132,
|
| 238 |
+
"mean_token_accuracy": 1.0,
|
| 239 |
+
"num_tokens": 97024.0,
|
| 240 |
+
"step": 48
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"entropy": 1.2649919390678406,
|
| 244 |
+
"epoch": 1.5625,
|
| 245 |
+
"grad_norm": 0.289153516292572,
|
| 246 |
+
"learning_rate": 7.03125e-05,
|
| 247 |
+
"loss": 0.2158,
|
| 248 |
+
"mean_token_accuracy": 1.0,
|
| 249 |
+
"num_tokens": 101120.0,
|
| 250 |
+
"step": 50
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"entropy": 0.9955946207046509,
|
| 254 |
+
"epoch": 1.625,
|
| 255 |
+
"grad_norm": 0.2607753276824951,
|
| 256 |
+
"learning_rate": 6.09375e-05,
|
| 257 |
+
"loss": 0.1913,
|
| 258 |
+
"mean_token_accuracy": 0.9997549057006836,
|
| 259 |
+
"num_tokens": 105216.0,
|
| 260 |
+
"step": 52
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"entropy": 1.3506205081939697,
|
| 264 |
+
"epoch": 1.6875,
|
| 265 |
+
"grad_norm": 0.2850724458694458,
|
| 266 |
+
"learning_rate": 5.156249999999999e-05,
|
| 267 |
+
"loss": 0.2234,
|
| 268 |
+
"mean_token_accuracy": 1.0,
|
| 269 |
+
"num_tokens": 109312.0,
|
| 270 |
+
"step": 54
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"entropy": 1.3460099697113037,
|
| 274 |
+
"epoch": 1.75,
|
| 275 |
+
"grad_norm": 0.23587484657764435,
|
| 276 |
+
"learning_rate": 4.2187499999999995e-05,
|
| 277 |
+
"loss": 0.2544,
|
| 278 |
+
"mean_token_accuracy": 1.0,
|
| 279 |
+
"num_tokens": 113408.0,
|
| 280 |
+
"step": 56
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"entropy": 1.4365423321723938,
|
| 284 |
+
"epoch": 1.8125,
|
| 285 |
+
"grad_norm": 0.3239842653274536,
|
| 286 |
+
"learning_rate": 3.28125e-05,
|
| 287 |
+
"loss": 0.2958,
|
| 288 |
+
"mean_token_accuracy": 1.0,
|
| 289 |
+
"num_tokens": 117504.0,
|
| 290 |
+
"step": 58
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"entropy": 1.567048728466034,
|
| 294 |
+
"epoch": 1.875,
|
| 295 |
+
"grad_norm": 0.34480002522468567,
|
| 296 |
+
"learning_rate": 2.3437499999999997e-05,
|
| 297 |
+
"loss": 0.3122,
|
| 298 |
+
"mean_token_accuracy": 1.0,
|
| 299 |
+
"num_tokens": 121600.0,
|
| 300 |
+
"step": 60
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"entropy": 1.344693124294281,
|
| 304 |
+
"epoch": 1.9375,
|
| 305 |
+
"grad_norm": 0.25280237197875977,
|
| 306 |
+
"learning_rate": 1.40625e-05,
|
| 307 |
+
"loss": 0.2472,
|
| 308 |
+
"mean_token_accuracy": 1.0,
|
| 309 |
+
"num_tokens": 125696.0,
|
| 310 |
+
"step": 62
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"entropy": 1.2893942594528198,
|
| 314 |
+
"epoch": 2.0,
|
| 315 |
+
"grad_norm": 0.5825140476226807,
|
| 316 |
+
"learning_rate": 4.6875e-06,
|
| 317 |
+
"loss": 0.2222,
|
| 318 |
+
"mean_token_accuracy": 1.0,
|
| 319 |
+
"num_tokens": 128512.0,
|
| 320 |
+
"step": 64
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 2.0,
|
| 324 |
+
"step": 64,
|
| 325 |
+
"total_flos": 59560143504384.0,
|
| 326 |
+
"train_loss": 0.32000101869925857,
|
| 327 |
+
"train_runtime": 720.302,
|
| 328 |
+
"train_samples_per_second": 0.711,
|
| 329 |
+
"train_steps_per_second": 0.089
|
| 330 |
+
}
|
| 331 |
+
]
|
events.out.tfevents.1758523788.ed35ea831684.8365.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:21a0268fbab3fb572425abfb4366a2e110c12add72177ce0d97fc045a4db0586
|
| 3 |
+
size 18423
|