data-ladder: dl_smollm2_135m_c4
Browse files
data-ladder-runs/dl_smollm2_135m_c4/metrics.csv
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
step,tokens,timed_hours,gpu_hours,train_loss_ema,neutral_bpb,ownval_bpb,lr,wallclock_s
|
| 2 |
+
17,8912896,0.001189,0.009515,10.2442,3.344837,2.723544,3.02e-04,41.6
|
| 3 |
+
23,12058624,0.002181,0.017445,9.6596,3.100666,2.507343,4.08e-04,76.3
|
| 4 |
+
30,15728640,0.003438,0.027508,9.0167,3.017381,2.411986,5.33e-04,111.9
|
| 5 |
+
40,20971520,0.005167,0.041332,8.3388,2.966188,2.347507,7.10e-04,148.9
|
| 6 |
+
54,28311552,0.007624,0.060994,7.6718,2.824540,2.249272,9.59e-04,188.9
|
| 7 |
+
71,37224448,0.010529,0.084230,7.1194,2.708298,2.132351,1.26e-03,230.3
|
| 8 |
+
95,49807360,0.014619,0.116949,6.6151,2.616458,2.046311,1.69e-03,276.2
|
| 9 |
+
127,66584576,0.020057,0.160454,6.2076,2.524731,1.953402,2.25e-03,326.8
|
| 10 |
+
169,88604672,0.027026,0.216206,5.8876,2.464635,1.878722,3.00e-03,382.9
|
| 11 |
+
226,118489088,0.036210,0.289679,5.4671,2.274580,1.724910,3.00e-03,447.1
|
| 12 |
+
301,157810688,0.048281,0.386246,4.9733,2.099514,1.580994,3.00e-03,521.5
|
| 13 |
+
401,210239488,0.063976,0.511804,4.4486,1.806137,1.420980,3.00e-03,609.0
|
| 14 |
+
535,280494080,0.084522,0.676173,4.1031,1.670054,1.329756,3.00e-03,714.1
|
| 15 |
+
713,373817344,0.111346,0.890767,3.8934,1.573025,1.262907,3.00e-03,841.9
|
| 16 |
+
951,498597888,0.146882,1.175060,3.7227,1.514364,1.215562,3.00e-03,1001.3
|
| 17 |
+
1269,665321472,0.193685,1.549477,3.6087,1.471878,1.178755,3.00e-03,1201.0
|
| 18 |
+
1692,887095296,0.254726,2.037812,3.5225,1.425927,1.148055,3.00e-03,1452.6
|
| 19 |
+
2256,1182793728,0.336031,2.688245,3.4401,1.399561,1.122619,3.00e-03,1776.8
|
| 20 |
+
3009,1577582592,0.444267,3.554136,3.3693,1.382324,1.100954,3.00e-03,2197.4
|
| 21 |
+
4012,2103443456,0.588316,4.706525,3.3179,1.360361,1.082374,3.00e-03,2746.9
|
| 22 |
+
5350,2804940800,0.780423,6.243388,3.2695,1.345031,1.065962,3.00e-03,3469.8
|
| 23 |
+
7135,3740794880,1.036757,8.294058,3.2286,1.330678,1.051906,3.00e-03,4423.7
|
| 24 |
+
9514,4988076032,1.378312,11.026494,3.1810,1.319429,1.039616,3.00e-03,5684.4
|
| 25 |
+
12687,6651641856,1.833732,14.669857,3.1518,1.310026,1.028940,3.00e-03,7355.1
|
| 26 |
+
15227,7983333376,2.198211,17.585686,3.0887,1.290210,1.009435,1.52e-03,8698.5
|
| 27 |
+
15565,8160542720,2.246708,17.973662,3.0779,1.287727,1.005844,1.21e-03,8904.3
|
| 28 |
+
15904,8338276352,2.295375,18.363002,3.0569,1.282406,1.002514,9.10e-04,9110.6
|
| 29 |
+
16242,8515485696,2.343868,18.750948,3.0576,1.280927,0.999313,6.07e-04,9316.0
|
| 30 |
+
16581,8693219328,2.392507,19.140056,3.0465,1.279166,0.996469,3.04e-04,9522.0
|
| 31 |
+
16919,8870428672,2.441002,19.528012,3.0261,1.278281,0.994662,8.96e-07,9727.5
|