| { | |
| "best_metric": 1.0839496850967407, | |
| "best_model_checkpoint": "data/Llama-31-8B_task-2_60-samples_config-4_full/checkpoint-267", | |
| "epoch": 99.82608695652173, | |
| "eval_steps": 500, | |
| "global_step": 287, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 0.5254858136177063, | |
| "learning_rate": 3.3333333333333335e-07, | |
| "loss": 1.6001, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 0.48315590620040894, | |
| "learning_rate": 6.666666666666667e-07, | |
| "loss": 1.5658, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "eval_loss": 1.5853713750839233, | |
| "eval_runtime": 12.9778, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 0.4942804276943207, | |
| "learning_rate": 1.3333333333333334e-06, | |
| "loss": 1.5728, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "eval_loss": 1.583631992340088, | |
| "eval_runtime": 12.9766, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 0.5288798809051514, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 1.5934, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 0.5049395561218262, | |
| "learning_rate": 2.666666666666667e-06, | |
| "loss": 1.583, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "eval_loss": 1.580271601676941, | |
| "eval_runtime": 12.9795, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 0.468879371881485, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 1.562, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "eval_loss": 1.5752835273742676, | |
| "eval_runtime": 12.9747, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 0.45035144686698914, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.5779, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 0.44937947392463684, | |
| "learning_rate": 4.666666666666667e-06, | |
| "loss": 1.5687, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 1.5688363313674927, | |
| "eval_runtime": 12.9875, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 0.439909964799881, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 1.5495, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 1.5600029230117798, | |
| "eval_runtime": 12.9772, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 0.4658964276313782, | |
| "learning_rate": 6e-06, | |
| "loss": 1.5541, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 0.5037180185317993, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 1.5493, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 1.548214316368103, | |
| "eval_runtime": 12.9784, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 0.47590604424476624, | |
| "learning_rate": 7.333333333333333e-06, | |
| "loss": 1.5379, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.5340198278427124, | |
| "eval_runtime": 12.9772, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 0.4758809506893158, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.5155, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "eval_loss": 1.5222147703170776, | |
| "eval_runtime": 12.9749, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 0.44333091378211975, | |
| "learning_rate": 8.666666666666668e-06, | |
| "loss": 1.5287, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 0.4040922224521637, | |
| "learning_rate": 9.333333333333334e-06, | |
| "loss": 1.5131, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "eval_loss": 1.5057040452957153, | |
| "eval_runtime": 12.977, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 0.39793452620506287, | |
| "learning_rate": 1e-05, | |
| "loss": 1.4971, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "eval_loss": 1.4859318733215332, | |
| "eval_runtime": 12.9807, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 11.130434782608695, | |
| "grad_norm": 0.4213944971561432, | |
| "learning_rate": 9.99864620589731e-06, | |
| "loss": 1.4727, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "grad_norm": 0.4004620909690857, | |
| "learning_rate": 9.994585556692624e-06, | |
| "loss": 1.4675, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "eval_loss": 1.4652339220046997, | |
| "eval_runtime": 12.9758, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 12.521739130434783, | |
| "grad_norm": 0.35341501235961914, | |
| "learning_rate": 9.987820251299121e-06, | |
| "loss": 1.4518, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "eval_loss": 1.4474080801010132, | |
| "eval_runtime": 12.9771, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 13.217391304347826, | |
| "grad_norm": 0.31634095311164856, | |
| "learning_rate": 9.978353953249023e-06, | |
| "loss": 1.4225, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "grad_norm": 0.32467326521873474, | |
| "learning_rate": 9.966191788709716e-06, | |
| "loss": 1.4267, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "eval_loss": 1.4301291704177856, | |
| "eval_runtime": 12.9757, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 14.608695652173914, | |
| "grad_norm": 0.3214071989059448, | |
| "learning_rate": 9.951340343707852e-06, | |
| "loss": 1.4004, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 14.956521739130435, | |
| "eval_loss": 1.413205623626709, | |
| "eval_runtime": 12.9784, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 15.304347826086957, | |
| "grad_norm": 0.26766690611839294, | |
| "learning_rate": 9.933807660562898e-06, | |
| "loss": 1.3894, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.24698252975940704, | |
| "learning_rate": 9.913603233532067e-06, | |
| "loss": 1.3993, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 1.3975735902786255, | |
| "eval_runtime": 12.9795, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 16.695652173913043, | |
| "grad_norm": 0.2775205671787262, | |
| "learning_rate": 9.890738003669029e-06, | |
| "loss": 1.3748, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 16.695652173913043, | |
| "eval_loss": 1.3880764245986938, | |
| "eval_runtime": 12.9755, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 17.391304347826086, | |
| "grad_norm": 0.2671767473220825, | |
| "learning_rate": 9.86522435289912e-06, | |
| "loss": 1.3664, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 17.73913043478261, | |
| "eval_loss": 1.3743128776550293, | |
| "eval_runtime": 12.98, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 18.08695652173913, | |
| "grad_norm": 0.23573799431324005, | |
| "learning_rate": 9.83707609731432e-06, | |
| "loss": 1.3404, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 18.782608695652176, | |
| "grad_norm": 0.23537954688072205, | |
| "learning_rate": 9.806308479691595e-06, | |
| "loss": 1.3465, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 18.782608695652176, | |
| "eval_loss": 1.361410140991211, | |
| "eval_runtime": 12.9811, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 19.47826086956522, | |
| "grad_norm": 0.24991613626480103, | |
| "learning_rate": 9.77293816123866e-06, | |
| "loss": 1.3407, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 19.82608695652174, | |
| "eval_loss": 1.3488320112228394, | |
| "eval_runtime": 12.9784, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 20.17391304347826, | |
| "grad_norm": 0.22332234680652618, | |
| "learning_rate": 9.736983212571646e-06, | |
| "loss": 1.3237, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 20.869565217391305, | |
| "grad_norm": 0.225299671292305, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 1.32, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 20.869565217391305, | |
| "eval_loss": 1.3369255065917969, | |
| "eval_runtime": 12.9791, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 21.565217391304348, | |
| "grad_norm": 0.24724063277244568, | |
| "learning_rate": 9.657398694630713e-06, | |
| "loss": 1.305, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 21.91304347826087, | |
| "eval_loss": 1.3247287273406982, | |
| "eval_runtime": 12.9775, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 22.26086956521739, | |
| "grad_norm": 0.24492307007312775, | |
| "learning_rate": 9.613812221777212e-06, | |
| "loss": 1.2923, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 22.956521739130434, | |
| "grad_norm": 0.2448989301919937, | |
| "learning_rate": 9.567727288213005e-06, | |
| "loss": 1.281, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 22.956521739130434, | |
| "eval_loss": 1.3119220733642578, | |
| "eval_runtime": 12.9822, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 23.652173913043477, | |
| "grad_norm": 0.22922120988368988, | |
| "learning_rate": 9.519168849742603e-06, | |
| "loss": 1.2869, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 1.298582911491394, | |
| "eval_runtime": 12.9756, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 24.347826086956523, | |
| "grad_norm": 0.20991288125514984, | |
| "learning_rate": 9.468163201617063e-06, | |
| "loss": 1.2523, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 24.695652173913043, | |
| "eval_loss": 1.2903255224227905, | |
| "eval_runtime": 12.978, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 25.043478260869566, | |
| "grad_norm": 0.22222064435482025, | |
| "learning_rate": 9.414737964294636e-06, | |
| "loss": 1.2628, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 25.73913043478261, | |
| "grad_norm": 0.21490232646465302, | |
| "learning_rate": 9.358922068483813e-06, | |
| "loss": 1.2642, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 25.73913043478261, | |
| "eval_loss": 1.278277039527893, | |
| "eval_runtime": 12.9804, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 26.434782608695652, | |
| "grad_norm": 0.22001975774765015, | |
| "learning_rate": 9.30074573947683e-06, | |
| "loss": 1.2323, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 26.782608695652176, | |
| "eval_loss": 1.265702247619629, | |
| "eval_runtime": 12.9772, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 27.130434782608695, | |
| "grad_norm": 0.20034939050674438, | |
| "learning_rate": 9.24024048078213e-06, | |
| "loss": 1.2395, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 27.82608695652174, | |
| "grad_norm": 0.19895507395267487, | |
| "learning_rate": 9.177439057064684e-06, | |
| "loss": 1.2121, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 27.82608695652174, | |
| "eval_loss": 1.253515362739563, | |
| "eval_runtime": 12.9793, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 28.52173913043478, | |
| "grad_norm": 0.19117023050785065, | |
| "learning_rate": 9.112375476403313e-06, | |
| "loss": 1.1896, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 28.869565217391305, | |
| "eval_loss": 1.2409998178482056, | |
| "eval_runtime": 12.9782, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 29.217391304347824, | |
| "grad_norm": 0.2141137570142746, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 1.2215, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 29.91304347826087, | |
| "grad_norm": 0.23155109584331512, | |
| "learning_rate": 8.97560398247424e-06, | |
| "loss": 1.1678, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 29.91304347826087, | |
| "eval_loss": 1.2283284664154053, | |
| "eval_runtime": 12.9808, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 30.608695652173914, | |
| "grad_norm": 0.2200462818145752, | |
| "learning_rate": 8.903970133383297e-06, | |
| "loss": 1.1768, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 30.956521739130434, | |
| "eval_loss": 1.2153583765029907, | |
| "eval_runtime": 12.9811, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 31.304347826086957, | |
| "grad_norm": 0.29315873980522156, | |
| "learning_rate": 8.83022221559489e-06, | |
| "loss": 1.1623, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "grad_norm": 0.22663454711437225, | |
| "learning_rate": 8.754400164907496e-06, | |
| "loss": 1.1824, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_loss": 1.20297372341156, | |
| "eval_runtime": 12.9813, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 32.69565217391305, | |
| "grad_norm": 0.2229231745004654, | |
| "learning_rate": 8.676545040299145e-06, | |
| "loss": 1.1589, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 32.69565217391305, | |
| "eval_loss": 1.1947580575942993, | |
| "eval_runtime": 12.9744, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 33.391304347826086, | |
| "grad_norm": 0.2454340010881424, | |
| "learning_rate": 8.596699001693257e-06, | |
| "loss": 1.126, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 33.73913043478261, | |
| "eval_loss": 1.1819590330123901, | |
| "eval_runtime": 12.9877, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 34.08695652173913, | |
| "grad_norm": 0.23333889245986938, | |
| "learning_rate": 8.51490528712831e-06, | |
| "loss": 1.1243, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 34.78260869565217, | |
| "grad_norm": 0.23953379690647125, | |
| "learning_rate": 8.43120818934367e-06, | |
| "loss": 1.1059, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 34.78260869565217, | |
| "eval_loss": 1.169416069984436, | |
| "eval_runtime": 12.9903, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 35.47826086956522, | |
| "grad_norm": 0.23915378749370575, | |
| "learning_rate": 8.345653031794292e-06, | |
| "loss": 1.1334, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 35.82608695652174, | |
| "eval_loss": 1.1582151651382446, | |
| "eval_runtime": 12.9858, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 36.17391304347826, | |
| "grad_norm": 0.23581968247890472, | |
| "learning_rate": 8.258286144107277e-06, | |
| "loss": 1.0911, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 36.869565217391305, | |
| "grad_norm": 0.207149475812912, | |
| "learning_rate": 8.16915483699355e-06, | |
| "loss": 1.1081, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 36.869565217391305, | |
| "eval_loss": 1.1483094692230225, | |
| "eval_runtime": 12.9841, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 37.56521739130435, | |
| "grad_norm": 0.20900563895702362, | |
| "learning_rate": 8.078307376628292e-06, | |
| "loss": 1.0794, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 37.91304347826087, | |
| "eval_loss": 1.139157772064209, | |
| "eval_runtime": 12.9828, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 38.26086956521739, | |
| "grad_norm": 0.19496272504329681, | |
| "learning_rate": 7.985792958513932e-06, | |
| "loss": 1.0814, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 38.95652173913044, | |
| "grad_norm": 0.20283730328083038, | |
| "learning_rate": 7.891661680839932e-06, | |
| "loss": 1.0614, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 38.95652173913044, | |
| "eval_loss": 1.1314579248428345, | |
| "eval_runtime": 13.0034, | |
| "eval_samples_per_second": 0.923, | |
| "eval_steps_per_second": 0.923, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 39.65217391304348, | |
| "grad_norm": 0.18000271916389465, | |
| "learning_rate": 7.795964517353734e-06, | |
| "loss": 1.0877, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 1.1258946657180786, | |
| "eval_runtime": 12.9887, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 40.34782608695652, | |
| "grad_norm": 0.1749521791934967, | |
| "learning_rate": 7.698753289757565e-06, | |
| "loss": 1.0198, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 40.69565217391305, | |
| "eval_loss": 1.122938632965088, | |
| "eval_runtime": 12.988, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 41.04347826086956, | |
| "grad_norm": 0.16494785249233246, | |
| "learning_rate": 7.600080639646077e-06, | |
| "loss": 1.0862, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 41.73913043478261, | |
| "grad_norm": 0.1447688788175583, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 1.0538, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 41.73913043478261, | |
| "eval_loss": 1.1193265914916992, | |
| "eval_runtime": 12.9856, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 42.43478260869565, | |
| "grad_norm": 0.1363561749458313, | |
| "learning_rate": 7.398565566251232e-06, | |
| "loss": 1.0351, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 42.78260869565217, | |
| "eval_loss": 1.116489291191101, | |
| "eval_runtime": 12.9859, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 43.130434782608695, | |
| "grad_norm": 0.149055615067482, | |
| "learning_rate": 7.295832266935059e-06, | |
| "loss": 1.0684, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 43.82608695652174, | |
| "grad_norm": 0.12285248190164566, | |
| "learning_rate": 7.191855733945388e-06, | |
| "loss": 1.0121, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 43.82608695652174, | |
| "eval_loss": 1.1143923997879028, | |
| "eval_runtime": 12.9844, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 44.52173913043478, | |
| "grad_norm": 0.13907888531684875, | |
| "learning_rate": 7.08669227240909e-06, | |
| "loss": 1.0475, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 44.869565217391305, | |
| "eval_loss": 1.1124660968780518, | |
| "eval_runtime": 12.9826, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 45.21739130434783, | |
| "grad_norm": 0.12828341126441956, | |
| "learning_rate": 6.980398830195785e-06, | |
| "loss": 1.0647, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 45.91304347826087, | |
| "grad_norm": 0.11993666738271713, | |
| "learning_rate": 6.873032967079562e-06, | |
| "loss": 1.035, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 45.91304347826087, | |
| "eval_loss": 1.1105235815048218, | |
| "eval_runtime": 12.9835, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 46.608695652173914, | |
| "grad_norm": 0.11968977749347687, | |
| "learning_rate": 6.7646528235693445e-06, | |
| "loss": 1.0582, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 46.95652173913044, | |
| "eval_loss": 1.1090105772018433, | |
| "eval_runtime": 12.981, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 47.30434782608695, | |
| "grad_norm": 0.12057274580001831, | |
| "learning_rate": 6.655317089424791e-06, | |
| "loss": 1.0257, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "grad_norm": 0.13167637586593628, | |
| "learning_rate": 6.545084971874738e-06, | |
| "loss": 1.029, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_loss": 1.10718834400177, | |
| "eval_runtime": 12.9849, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 48.69565217391305, | |
| "grad_norm": 0.11730153858661652, | |
| "learning_rate": 6.434016163555452e-06, | |
| "loss": 1.0353, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 48.69565217391305, | |
| "eval_loss": 1.1063708066940308, | |
| "eval_runtime": 12.9872, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 49.391304347826086, | |
| "grad_norm": 0.1166403666138649, | |
| "learning_rate": 6.322170810186013e-06, | |
| "loss": 1.0203, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 49.73913043478261, | |
| "eval_loss": 1.1048336029052734, | |
| "eval_runtime": 12.9919, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 50.08695652173913, | |
| "grad_norm": 0.13392803072929382, | |
| "learning_rate": 6.209609477998339e-06, | |
| "loss": 1.0411, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 50.78260869565217, | |
| "grad_norm": 0.11808019131422043, | |
| "learning_rate": 6.0963931209395165e-06, | |
| "loss": 1.0313, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 50.78260869565217, | |
| "eval_loss": 1.1034945249557495, | |
| "eval_runtime": 12.9911, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 51.47826086956522, | |
| "grad_norm": 0.12019839882850647, | |
| "learning_rate": 5.982583047664151e-06, | |
| "loss": 1.0473, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 51.82608695652174, | |
| "eval_loss": 1.1025954484939575, | |
| "eval_runtime": 12.987, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 149 | |
| }, | |
| { | |
| "epoch": 52.17391304347826, | |
| "grad_norm": 0.13823947310447693, | |
| "learning_rate": 5.8682408883346535e-06, | |
| "loss": 1.0099, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 52.869565217391305, | |
| "grad_norm": 0.13493500649929047, | |
| "learning_rate": 5.753428561247416e-06, | |
| "loss": 1.0189, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 52.869565217391305, | |
| "eval_loss": 1.1011220216751099, | |
| "eval_runtime": 12.9858, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 53.56521739130435, | |
| "grad_norm": 0.11930125206708908, | |
| "learning_rate": 5.638208239302975e-06, | |
| "loss": 1.0088, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 53.91304347826087, | |
| "eval_loss": 1.100091576576233, | |
| "eval_runtime": 12.9925, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 54.26086956521739, | |
| "grad_norm": 0.1283315271139145, | |
| "learning_rate": 5.522642316338268e-06, | |
| "loss": 1.0471, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 54.95652173913044, | |
| "grad_norm": 0.12327709794044495, | |
| "learning_rate": 5.406793373339292e-06, | |
| "loss": 1.0336, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 54.95652173913044, | |
| "eval_loss": 1.0989197492599487, | |
| "eval_runtime": 12.9836, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 55.65217391304348, | |
| "grad_norm": 0.1234331727027893, | |
| "learning_rate": 5.290724144552379e-06, | |
| "loss": 1.0014, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_loss": 1.098119854927063, | |
| "eval_runtime": 12.9883, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 56.34782608695652, | |
| "grad_norm": 0.1393013894557953, | |
| "learning_rate": 5.174497483512506e-06, | |
| "loss": 1.0036, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 56.69565217391305, | |
| "eval_loss": 1.0972312688827515, | |
| "eval_runtime": 12.9843, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 163 | |
| }, | |
| { | |
| "epoch": 57.04347826086956, | |
| "grad_norm": 0.12967315316200256, | |
| "learning_rate": 5.0581763290069865e-06, | |
| "loss": 1.0466, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 57.73913043478261, | |
| "grad_norm": 0.1203417107462883, | |
| "learning_rate": 4.941823670993016e-06, | |
| "loss": 1.0266, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 57.73913043478261, | |
| "eval_loss": 1.0962202548980713, | |
| "eval_runtime": 12.984, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 58.43478260869565, | |
| "grad_norm": 0.12664881348609924, | |
| "learning_rate": 4.825502516487497e-06, | |
| "loss": 0.9893, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 58.78260869565217, | |
| "eval_loss": 1.0956058502197266, | |
| "eval_runtime": 12.983, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 169 | |
| }, | |
| { | |
| "epoch": 59.130434782608695, | |
| "grad_norm": 0.1268896907567978, | |
| "learning_rate": 4.7092758554476215e-06, | |
| "loss": 1.0363, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 59.82608695652174, | |
| "grad_norm": 0.12260066717863083, | |
| "learning_rate": 4.59320662666071e-06, | |
| "loss": 1.0122, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 59.82608695652174, | |
| "eval_loss": 1.0947729349136353, | |
| "eval_runtime": 12.9887, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 60.52173913043478, | |
| "grad_norm": 0.12348895519971848, | |
| "learning_rate": 4.477357683661734e-06, | |
| "loss": 1.0456, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 60.869565217391305, | |
| "eval_loss": 1.0938516855239868, | |
| "eval_runtime": 12.9852, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 61.21739130434783, | |
| "grad_norm": 0.1181364431977272, | |
| "learning_rate": 4.361791760697027e-06, | |
| "loss": 1.0021, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 61.91304347826087, | |
| "grad_norm": 0.12426912784576416, | |
| "learning_rate": 4.246571438752585e-06, | |
| "loss": 0.9873, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 61.91304347826087, | |
| "eval_loss": 1.0932997465133667, | |
| "eval_runtime": 12.9851, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 62.608695652173914, | |
| "grad_norm": 0.12317227572202682, | |
| "learning_rate": 4.131759111665349e-06, | |
| "loss": 1.0189, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 62.95652173913044, | |
| "eval_loss": 1.0926170349121094, | |
| "eval_runtime": 12.9854, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 181 | |
| }, | |
| { | |
| "epoch": 63.30434782608695, | |
| "grad_norm": 0.14774170517921448, | |
| "learning_rate": 4.017416952335849e-06, | |
| "loss": 0.9927, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "grad_norm": 0.12388727813959122, | |
| "learning_rate": 3.903606879060483e-06, | |
| "loss": 1.0325, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_loss": 1.0917848348617554, | |
| "eval_runtime": 12.981, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 64.69565217391305, | |
| "grad_norm": 0.12087590992450714, | |
| "learning_rate": 3.790390522001662e-06, | |
| "loss": 1.0081, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 64.69565217391305, | |
| "eval_loss": 1.0912106037139893, | |
| "eval_runtime": 12.9845, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 65.3913043478261, | |
| "grad_norm": 0.12642121315002441, | |
| "learning_rate": 3.6778291898139907e-06, | |
| "loss": 0.995, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 65.73913043478261, | |
| "eval_loss": 1.0907906293869019, | |
| "eval_runtime": 12.9837, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 66.08695652173913, | |
| "grad_norm": 0.13900116086006165, | |
| "learning_rate": 3.5659838364445505e-06, | |
| "loss": 1.0352, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 66.78260869565217, | |
| "grad_norm": 0.13428539037704468, | |
| "learning_rate": 3.4549150281252635e-06, | |
| "loss": 1.0104, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 66.78260869565217, | |
| "eval_loss": 1.0903018712997437, | |
| "eval_runtime": 12.9828, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 67.47826086956522, | |
| "grad_norm": 0.12461841106414795, | |
| "learning_rate": 3.3446829105752103e-06, | |
| "loss": 0.9979, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 67.82608695652173, | |
| "eval_loss": 1.089614748954773, | |
| "eval_runtime": 12.9828, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 68.17391304347827, | |
| "grad_norm": 0.13673454523086548, | |
| "learning_rate": 3.2353471764306567e-06, | |
| "loss": 1.015, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 68.8695652173913, | |
| "grad_norm": 0.12500186264514923, | |
| "learning_rate": 3.12696703292044e-06, | |
| "loss": 0.9927, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 68.8695652173913, | |
| "eval_loss": 1.0892704725265503, | |
| "eval_runtime": 12.986, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 69.56521739130434, | |
| "grad_norm": 0.12714624404907227, | |
| "learning_rate": 3.019601169804216e-06, | |
| "loss": 0.9898, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 69.91304347826087, | |
| "eval_loss": 1.0887213945388794, | |
| "eval_runtime": 12.9821, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 70.26086956521739, | |
| "grad_norm": 0.1471695601940155, | |
| "learning_rate": 2.9133077275909112e-06, | |
| "loss": 1.0112, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 70.95652173913044, | |
| "grad_norm": 0.12886075675487518, | |
| "learning_rate": 2.8081442660546126e-06, | |
| "loss": 1.0087, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 70.95652173913044, | |
| "eval_loss": 1.0882389545440674, | |
| "eval_runtime": 12.9829, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 71.65217391304348, | |
| "grad_norm": 0.13256506621837616, | |
| "learning_rate": 2.7041677330649408e-06, | |
| "loss": 0.9903, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_loss": 1.0877516269683838, | |
| "eval_runtime": 12.9825, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 72.34782608695652, | |
| "grad_norm": 0.131199449300766, | |
| "learning_rate": 2.601434433748771e-06, | |
| "loss": 1.0198, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 72.69565217391305, | |
| "eval_loss": 1.0877233743667603, | |
| "eval_runtime": 12.9886, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 73.04347826086956, | |
| "grad_norm": 0.12681709229946136, | |
| "learning_rate": 2.5000000000000015e-06, | |
| "loss": 0.9942, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 73.73913043478261, | |
| "grad_norm": 0.12825633585453033, | |
| "learning_rate": 2.3999193603539234e-06, | |
| "loss": 1.0078, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 73.73913043478261, | |
| "eval_loss": 1.0873947143554688, | |
| "eval_runtime": 12.985, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 74.43478260869566, | |
| "grad_norm": 0.12875820696353912, | |
| "learning_rate": 2.3012467102424373e-06, | |
| "loss": 1.0056, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 74.78260869565217, | |
| "eval_loss": 1.086978793144226, | |
| "eval_runtime": 12.9911, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 75.1304347826087, | |
| "grad_norm": 0.16595281660556793, | |
| "learning_rate": 2.204035482646267e-06, | |
| "loss": 0.9857, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 75.82608695652173, | |
| "grad_norm": 0.1281590312719345, | |
| "learning_rate": 2.1083383191600676e-06, | |
| "loss": 1.0114, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 75.82608695652173, | |
| "eval_loss": 1.0867377519607544, | |
| "eval_runtime": 12.9863, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 76.52173913043478, | |
| "grad_norm": 0.1323501169681549, | |
| "learning_rate": 2.0142070414860704e-06, | |
| "loss": 0.9982, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 76.8695652173913, | |
| "eval_loss": 1.0864088535308838, | |
| "eval_runtime": 12.9841, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 221 | |
| }, | |
| { | |
| "epoch": 77.21739130434783, | |
| "grad_norm": 0.1450796127319336, | |
| "learning_rate": 1.9216926233717087e-06, | |
| "loss": 0.974, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 77.91304347826087, | |
| "grad_norm": 0.13413220643997192, | |
| "learning_rate": 1.8308451630064484e-06, | |
| "loss": 1.0105, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 77.91304347826087, | |
| "eval_loss": 1.086008906364441, | |
| "eval_runtime": 13.0104, | |
| "eval_samples_per_second": 0.922, | |
| "eval_steps_per_second": 0.922, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 78.6086956521739, | |
| "grad_norm": 0.1377415508031845, | |
| "learning_rate": 1.7417138558927244e-06, | |
| "loss": 1.0033, | |
| "step": 226 | |
| }, | |
| { | |
| "epoch": 78.95652173913044, | |
| "eval_loss": 1.085876226425171, | |
| "eval_runtime": 12.9858, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 227 | |
| }, | |
| { | |
| "epoch": 79.30434782608695, | |
| "grad_norm": 0.13374638557434082, | |
| "learning_rate": 1.6543469682057105e-06, | |
| "loss": 0.9876, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 0.12589867413043976, | |
| "learning_rate": 1.5687918106563326e-06, | |
| "loss": 1.0024, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_loss": 1.0857700109481812, | |
| "eval_runtime": 12.9853, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 80.69565217391305, | |
| "grad_norm": 0.1440684199333191, | |
| "learning_rate": 1.4850947128716914e-06, | |
| "loss": 1.0091, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 80.69565217391305, | |
| "eval_loss": 1.0854946374893188, | |
| "eval_runtime": 12.994, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 81.3913043478261, | |
| "grad_norm": 0.13353127241134644, | |
| "learning_rate": 1.4033009983067454e-06, | |
| "loss": 0.9971, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 81.73913043478261, | |
| "eval_loss": 1.0853348970413208, | |
| "eval_runtime": 12.9853, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 82.08695652173913, | |
| "grad_norm": 0.14419566094875336, | |
| "learning_rate": 1.3234549597008572e-06, | |
| "loss": 0.9801, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 82.78260869565217, | |
| "grad_norm": 0.13215622305870056, | |
| "learning_rate": 1.2455998350925042e-06, | |
| "loss": 0.969, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 82.78260869565217, | |
| "eval_loss": 1.0850697755813599, | |
| "eval_runtime": 12.9889, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 83.47826086956522, | |
| "grad_norm": 0.1365135759115219, | |
| "learning_rate": 1.1697777844051105e-06, | |
| "loss": 1.0242, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 83.82608695652173, | |
| "eval_loss": 1.0847272872924805, | |
| "eval_runtime": 12.9926, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 241 | |
| }, | |
| { | |
| "epoch": 84.17391304347827, | |
| "grad_norm": 0.14472560584545135, | |
| "learning_rate": 1.096029866616704e-06, | |
| "loss": 1.0026, | |
| "step": 242 | |
| }, | |
| { | |
| "epoch": 84.8695652173913, | |
| "grad_norm": 0.13619466125965118, | |
| "learning_rate": 1.0243960175257605e-06, | |
| "loss": 0.9949, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 84.8695652173913, | |
| "eval_loss": 1.0849597454071045, | |
| "eval_runtime": 12.9855, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 85.56521739130434, | |
| "grad_norm": 0.13221730291843414, | |
| "learning_rate": 9.549150281252633e-07, | |
| "loss": 0.9715, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 85.91304347826087, | |
| "eval_loss": 1.0846881866455078, | |
| "eval_runtime": 12.9911, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 247 | |
| }, | |
| { | |
| "epoch": 86.26086956521739, | |
| "grad_norm": 0.14602892100811005, | |
| "learning_rate": 8.876245235966884e-07, | |
| "loss": 0.9918, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 86.95652173913044, | |
| "grad_norm": 0.1370145082473755, | |
| "learning_rate": 8.225609429353187e-07, | |
| "loss": 1.0164, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 86.95652173913044, | |
| "eval_loss": 1.084583044052124, | |
| "eval_runtime": 12.9873, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 87.65217391304348, | |
| "grad_norm": 0.1295238435268402, | |
| "learning_rate": 7.597595192178702e-07, | |
| "loss": 0.9729, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_loss": 1.084546685218811, | |
| "eval_runtime": 12.9858, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 88.34782608695652, | |
| "grad_norm": 0.12774939835071564, | |
| "learning_rate": 6.992542605231739e-07, | |
| "loss": 1.0065, | |
| "step": 254 | |
| }, | |
| { | |
| "epoch": 88.69565217391305, | |
| "eval_loss": 1.0845379829406738, | |
| "eval_runtime": 12.9859, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 89.04347826086956, | |
| "grad_norm": 0.12875907123088837, | |
| "learning_rate": 6.410779315161885e-07, | |
| "loss": 1.0004, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 89.73913043478261, | |
| "grad_norm": 0.13160941004753113, | |
| "learning_rate": 5.852620357053651e-07, | |
| "loss": 0.994, | |
| "step": 258 | |
| }, | |
| { | |
| "epoch": 89.73913043478261, | |
| "eval_loss": 1.0845232009887695, | |
| "eval_runtime": 12.9865, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 258 | |
| }, | |
| { | |
| "epoch": 90.43478260869566, | |
| "grad_norm": 0.1342068612575531, | |
| "learning_rate": 5.318367983829393e-07, | |
| "loss": 0.9852, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 90.78260869565217, | |
| "eval_loss": 1.084337830543518, | |
| "eval_runtime": 12.9887, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 261 | |
| }, | |
| { | |
| "epoch": 91.1304347826087, | |
| "grad_norm": 0.14868062734603882, | |
| "learning_rate": 4.808311502573976e-07, | |
| "loss": 1.0154, | |
| "step": 262 | |
| }, | |
| { | |
| "epoch": 91.82608695652173, | |
| "grad_norm": 0.14731329679489136, | |
| "learning_rate": 4.322727117869951e-07, | |
| "loss": 0.9755, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 91.82608695652173, | |
| "eval_loss": 1.0842256546020508, | |
| "eval_runtime": 12.9863, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 92.52173913043478, | |
| "grad_norm": 0.14365264773368835, | |
| "learning_rate": 3.8618777822278854e-07, | |
| "loss": 1.0191, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 92.8695652173913, | |
| "eval_loss": 1.0839496850967407, | |
| "eval_runtime": 12.9888, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 267 | |
| }, | |
| { | |
| "epoch": 93.21739130434783, | |
| "grad_norm": 0.15139809250831604, | |
| "learning_rate": 3.426013053692878e-07, | |
| "loss": 0.9765, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 93.91304347826087, | |
| "grad_norm": 0.14520440995693207, | |
| "learning_rate": 3.015368960704584e-07, | |
| "loss": 0.9864, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 93.91304347826087, | |
| "eval_loss": 1.0840731859207153, | |
| "eval_runtime": 12.9887, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 94.6086956521739, | |
| "grad_norm": 0.13488347828388214, | |
| "learning_rate": 2.63016787428354e-07, | |
| "loss": 0.9773, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 94.95652173913044, | |
| "eval_loss": 1.084143042564392, | |
| "eval_runtime": 12.9839, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 95.30434782608695, | |
| "grad_norm": 0.14086610078811646, | |
| "learning_rate": 2.2706183876134047e-07, | |
| "loss": 1.0173, | |
| "step": 274 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "grad_norm": 0.13791334629058838, | |
| "learning_rate": 1.9369152030840553e-07, | |
| "loss": 0.9869, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_loss": 1.0842411518096924, | |
| "eval_runtime": 12.9864, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 96.69565217391305, | |
| "grad_norm": 0.14270715415477753, | |
| "learning_rate": 1.6292390268568103e-07, | |
| "loss": 0.986, | |
| "step": 278 | |
| }, | |
| { | |
| "epoch": 96.69565217391305, | |
| "eval_loss": 1.0841034650802612, | |
| "eval_runtime": 12.985, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 278 | |
| }, | |
| { | |
| "epoch": 97.3913043478261, | |
| "grad_norm": 0.13356775045394897, | |
| "learning_rate": 1.3477564710088097e-07, | |
| "loss": 0.9925, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 97.73913043478261, | |
| "eval_loss": 1.084011197090149, | |
| "eval_runtime": 12.9842, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 281 | |
| }, | |
| { | |
| "epoch": 98.08695652173913, | |
| "grad_norm": 0.15943579375743866, | |
| "learning_rate": 1.0926199633097156e-07, | |
| "loss": 0.9947, | |
| "step": 282 | |
| }, | |
| { | |
| "epoch": 98.78260869565217, | |
| "grad_norm": 0.1478174775838852, | |
| "learning_rate": 8.639676646793382e-08, | |
| "loss": 0.9882, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 98.78260869565217, | |
| "eval_loss": 1.0840474367141724, | |
| "eval_runtime": 12.9842, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 99.47826086956522, | |
| "grad_norm": 0.14343644678592682, | |
| "learning_rate": 6.61923394371039e-08, | |
| "loss": 0.9917, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 99.82608695652173, | |
| "eval_loss": 1.0840203762054443, | |
| "eval_runtime": 12.9837, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 287 | |
| }, | |
| { | |
| "epoch": 99.82608695652173, | |
| "step": 287, | |
| "total_flos": 4.445754122074849e+17, | |
| "train_loss": 1.1405799933011525, | |
| "train_runtime": 16000.7915, | |
| "train_samples_per_second": 0.431, | |
| "train_steps_per_second": 0.019 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 300, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 150, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 7, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.445754122074849e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |