| { | |
| "best_metric": 0.46568557620048523, | |
| "best_model_checkpoint": "data/Llama-31-8B_task-3_60-samples_config-3/checkpoint-149", | |
| "epoch": 32.869565217391305, | |
| "eval_steps": 500, | |
| "global_step": 189, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17391304347826086, | |
| "grad_norm": 3.831383466720581, | |
| "learning_rate": 1.3333333333333336e-07, | |
| "loss": 2.9855, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 1.815189242362976, | |
| "learning_rate": 2.666666666666667e-07, | |
| "loss": 2.0019, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 2.1689257621765137, | |
| "learning_rate": 5.333333333333335e-07, | |
| "loss": 2.4489, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "eval_loss": 2.470275402069092, | |
| "eval_runtime": 11.1851, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0434782608695652, | |
| "grad_norm": 2.6814849376678467, | |
| "learning_rate": 8.000000000000001e-07, | |
| "loss": 2.449, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 3.0052318572998047, | |
| "learning_rate": 1.066666666666667e-06, | |
| "loss": 2.4279, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "grad_norm": 3.0235185623168945, | |
| "learning_rate": 1.3333333333333334e-06, | |
| "loss": 2.4891, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.9130434782608696, | |
| "eval_loss": 2.4544930458068848, | |
| "eval_runtime": 11.1948, | |
| "eval_samples_per_second": 1.072, | |
| "eval_steps_per_second": 1.072, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 2.1276910305023193, | |
| "learning_rate": 1.6000000000000001e-06, | |
| "loss": 2.4166, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.4347826086956523, | |
| "grad_norm": 3.4439425468444824, | |
| "learning_rate": 1.8666666666666669e-06, | |
| "loss": 2.7413, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 2.9551491737365723, | |
| "learning_rate": 2.133333333333334e-06, | |
| "loss": 2.4697, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.9565217391304346, | |
| "eval_loss": 2.4171502590179443, | |
| "eval_runtime": 11.1863, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 3.130434782608696, | |
| "grad_norm": 2.11788010597229, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 1.9378, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 2.177309274673462, | |
| "learning_rate": 2.666666666666667e-06, | |
| "loss": 2.3009, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "grad_norm": 3.9081153869628906, | |
| "learning_rate": 2.9333333333333338e-06, | |
| "loss": 2.5196, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.3542613983154297, | |
| "eval_runtime": 11.1853, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 2.6722283363342285, | |
| "learning_rate": 3.2000000000000003e-06, | |
| "loss": 2.4047, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.521739130434782, | |
| "grad_norm": 2.0837063789367676, | |
| "learning_rate": 3.4666666666666672e-06, | |
| "loss": 2.1966, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 3.5176823139190674, | |
| "learning_rate": 3.7333333333333337e-06, | |
| "loss": 2.4481, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 2.2785379886627197, | |
| "eval_runtime": 11.1838, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.217391304347826, | |
| "grad_norm": 2.1525745391845703, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 2.1104, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 3.053701400756836, | |
| "learning_rate": 4.266666666666668e-06, | |
| "loss": 2.1746, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "grad_norm": 3.4729630947113037, | |
| "learning_rate": 4.533333333333334e-06, | |
| "loss": 2.2925, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 2.150887966156006, | |
| "eval_runtime": 11.186, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 3.08322811126709, | |
| "learning_rate": 4.800000000000001e-06, | |
| "loss": 2.1084, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.608695652173913, | |
| "grad_norm": 2.111081123352051, | |
| "learning_rate": 5.0666666666666676e-06, | |
| "loss": 2.0085, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 3.0863735675811768, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 2.0169, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 1.9727939367294312, | |
| "eval_runtime": 11.1859, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.304347826086957, | |
| "grad_norm": 2.0293078422546387, | |
| "learning_rate": 5.600000000000001e-06, | |
| "loss": 1.9966, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 2.927000045776367, | |
| "learning_rate": 5.8666666666666675e-06, | |
| "loss": 1.9564, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.686829924583435, | |
| "learning_rate": 6.133333333333334e-06, | |
| "loss": 1.6364, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.7440919876098633, | |
| "eval_runtime": 11.1911, | |
| "eval_samples_per_second": 1.072, | |
| "eval_steps_per_second": 1.072, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 2.4069645404815674, | |
| "learning_rate": 6.4000000000000006e-06, | |
| "loss": 1.7646, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "grad_norm": 2.646470546722412, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 1.6047, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.869565217391305, | |
| "eval_loss": 1.5191912651062012, | |
| "eval_runtime": 11.1878, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 2.2305424213409424, | |
| "learning_rate": 6.9333333333333344e-06, | |
| "loss": 1.4515, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.391304347826088, | |
| "grad_norm": 2.5355496406555176, | |
| "learning_rate": 7.2000000000000005e-06, | |
| "loss": 1.4093, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 2.0611608028411865, | |
| "learning_rate": 7.4666666666666675e-06, | |
| "loss": 1.3126, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 9.91304347826087, | |
| "eval_loss": 1.2352849245071411, | |
| "eval_runtime": 11.1877, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 10.08695652173913, | |
| "grad_norm": 1.9300768375396729, | |
| "learning_rate": 7.733333333333334e-06, | |
| "loss": 1.2497, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 1.9045050144195557, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.1055, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "grad_norm": 1.8269885778427124, | |
| "learning_rate": 8.266666666666667e-06, | |
| "loss": 1.0406, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 10.956521739130435, | |
| "eval_loss": 0.9674362540245056, | |
| "eval_runtime": 11.1886, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 11.130434782608695, | |
| "grad_norm": 1.7003505229949951, | |
| "learning_rate": 8.533333333333335e-06, | |
| "loss": 0.9995, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 11.478260869565217, | |
| "grad_norm": 2.517176866531372, | |
| "learning_rate": 8.8e-06, | |
| "loss": 0.7261, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "grad_norm": 1.7687137126922607, | |
| "learning_rate": 9.066666666666667e-06, | |
| "loss": 0.8254, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.7401664853096008, | |
| "eval_runtime": 11.1831, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 12.173913043478262, | |
| "grad_norm": 1.434043049812317, | |
| "learning_rate": 9.333333333333334e-06, | |
| "loss": 0.6623, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 12.521739130434783, | |
| "grad_norm": 1.1792118549346924, | |
| "learning_rate": 9.600000000000001e-06, | |
| "loss": 0.6543, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "grad_norm": 1.1398481130599976, | |
| "learning_rate": 9.866666666666668e-06, | |
| "loss": 0.7723, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "eval_loss": 0.6372489929199219, | |
| "eval_runtime": 11.1866, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 13.217391304347826, | |
| "grad_norm": 1.3324942588806152, | |
| "learning_rate": 9.999945845889795e-06, | |
| "loss": 0.5831, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 13.565217391304348, | |
| "grad_norm": 0.8719578981399536, | |
| "learning_rate": 9.999512620046523e-06, | |
| "loss": 0.4151, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "grad_norm": 0.5932775735855103, | |
| "learning_rate": 9.99864620589731e-06, | |
| "loss": 0.4327, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "eval_loss": 0.5750009417533875, | |
| "eval_runtime": 11.1885, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 14.26086956521739, | |
| "grad_norm": 0.9070873260498047, | |
| "learning_rate": 9.99734667851357e-06, | |
| "loss": 0.7031, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 14.608695652173914, | |
| "grad_norm": 0.7875546216964722, | |
| "learning_rate": 9.995614150494293e-06, | |
| "loss": 0.5401, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 14.956521739130435, | |
| "grad_norm": 0.7815837264060974, | |
| "learning_rate": 9.993448771956285e-06, | |
| "loss": 0.279, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 14.956521739130435, | |
| "eval_loss": 0.531262218952179, | |
| "eval_runtime": 11.1884, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 15.304347826086957, | |
| "grad_norm": 0.558903694152832, | |
| "learning_rate": 9.99085073052117e-06, | |
| "loss": 0.5445, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 15.652173913043478, | |
| "grad_norm": 0.8311617374420166, | |
| "learning_rate": 9.987820251299121e-06, | |
| "loss": 0.4926, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.7534677386283875, | |
| "learning_rate": 9.984357596869369e-06, | |
| "loss": 0.4039, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.5046530961990356, | |
| "eval_runtime": 11.1879, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 16.347826086956523, | |
| "grad_norm": 2.769418478012085, | |
| "learning_rate": 9.980463067257437e-06, | |
| "loss": 0.4303, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 16.695652173913043, | |
| "grad_norm": 0.5363882780075073, | |
| "learning_rate": 9.976136999909156e-06, | |
| "loss": 0.5227, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 16.869565217391305, | |
| "eval_loss": 0.48834142088890076, | |
| "eval_runtime": 11.1855, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 17.043478260869566, | |
| "grad_norm": 0.6151431202888489, | |
| "learning_rate": 9.971379769661422e-06, | |
| "loss": 0.3441, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 17.391304347826086, | |
| "grad_norm": 0.7451347708702087, | |
| "learning_rate": 9.966191788709716e-06, | |
| "loss": 0.502, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 17.73913043478261, | |
| "grad_norm": 0.5834351778030396, | |
| "learning_rate": 9.960573506572391e-06, | |
| "loss": 0.3704, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 17.91304347826087, | |
| "eval_loss": 0.4936355650424957, | |
| "eval_runtime": 11.1889, | |
| "eval_samples_per_second": 1.072, | |
| "eval_steps_per_second": 1.072, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 18.08695652173913, | |
| "grad_norm": 0.4578331410884857, | |
| "learning_rate": 9.95452541005172e-06, | |
| "loss": 0.2609, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 18.434782608695652, | |
| "grad_norm": 0.4481121003627777, | |
| "learning_rate": 9.948048023191728e-06, | |
| "loss": 0.3326, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 18.782608695652176, | |
| "grad_norm": 0.3353727459907532, | |
| "learning_rate": 9.941141907232766e-06, | |
| "loss": 0.497, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 18.956521739130434, | |
| "eval_loss": 0.49074581265449524, | |
| "eval_runtime": 11.187, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 19.130434782608695, | |
| "grad_norm": 0.4987470507621765, | |
| "learning_rate": 9.933807660562898e-06, | |
| "loss": 0.4857, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 19.47826086956522, | |
| "grad_norm": 0.42145681381225586, | |
| "learning_rate": 9.926045918666045e-06, | |
| "loss": 0.3594, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 19.82608695652174, | |
| "grad_norm": 0.2636556029319763, | |
| "learning_rate": 9.91785735406693e-06, | |
| "loss": 0.4428, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.4828261137008667, | |
| "eval_runtime": 11.1843, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 20.17391304347826, | |
| "grad_norm": 0.42601507902145386, | |
| "learning_rate": 9.909242676272797e-06, | |
| "loss": 0.3907, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 20.52173913043478, | |
| "grad_norm": 0.32994499802589417, | |
| "learning_rate": 9.90020263171194e-06, | |
| "loss": 0.4499, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 20.869565217391305, | |
| "grad_norm": 0.4801340699195862, | |
| "learning_rate": 9.890738003669029e-06, | |
| "loss": 0.3477, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 20.869565217391305, | |
| "eval_loss": 0.482466459274292, | |
| "eval_runtime": 11.187, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 21.217391304347824, | |
| "grad_norm": 0.32882365584373474, | |
| "learning_rate": 9.880849612217238e-06, | |
| "loss": 0.2927, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 21.565217391304348, | |
| "grad_norm": 0.3345174491405487, | |
| "learning_rate": 9.870538314147194e-06, | |
| "loss": 0.3384, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 21.91304347826087, | |
| "grad_norm": 0.4246956408023834, | |
| "learning_rate": 9.859805002892733e-06, | |
| "loss": 0.4628, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 21.91304347826087, | |
| "eval_loss": 0.4857006072998047, | |
| "eval_runtime": 11.1862, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 22.26086956521739, | |
| "grad_norm": 0.35164156556129456, | |
| "learning_rate": 9.84865060845349e-06, | |
| "loss": 0.3811, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 22.608695652173914, | |
| "grad_norm": 0.22983089089393616, | |
| "learning_rate": 9.83707609731432e-06, | |
| "loss": 0.3886, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 22.956521739130434, | |
| "grad_norm": 0.333185076713562, | |
| "learning_rate": 9.825082472361558e-06, | |
| "loss": 0.367, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 22.956521739130434, | |
| "eval_loss": 0.48269978165626526, | |
| "eval_runtime": 11.1827, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 23.304347826086957, | |
| "grad_norm": 0.3328111469745636, | |
| "learning_rate": 9.812670772796113e-06, | |
| "loss": 0.3419, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 23.652173913043477, | |
| "grad_norm": 0.3405808210372925, | |
| "learning_rate": 9.799842074043438e-06, | |
| "loss": 0.4314, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 0.44388675689697266, | |
| "learning_rate": 9.786597487660336e-06, | |
| "loss": 0.2597, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 0.4767860472202301, | |
| "eval_runtime": 11.184, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 24.347826086956523, | |
| "grad_norm": 0.18694591522216797, | |
| "learning_rate": 9.77293816123866e-06, | |
| "loss": 0.3109, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 24.695652173913043, | |
| "grad_norm": 0.44868358969688416, | |
| "learning_rate": 9.75886527830587e-06, | |
| "loss": 0.3473, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 24.869565217391305, | |
| "eval_loss": 0.46839889883995056, | |
| "eval_runtime": 11.1867, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 25.043478260869566, | |
| "grad_norm": 0.4045964777469635, | |
| "learning_rate": 9.744380058222483e-06, | |
| "loss": 0.4445, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 25.391304347826086, | |
| "grad_norm": 0.3353895843029022, | |
| "learning_rate": 9.729483756076436e-06, | |
| "loss": 0.3212, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 25.73913043478261, | |
| "grad_norm": 0.3463374078273773, | |
| "learning_rate": 9.714177662574316e-06, | |
| "loss": 0.3795, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 25.91304347826087, | |
| "eval_loss": 0.46568557620048523, | |
| "eval_runtime": 11.1848, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 149 | |
| }, | |
| { | |
| "epoch": 26.08695652173913, | |
| "grad_norm": 0.5068124532699585, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 0.3142, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 26.434782608695652, | |
| "grad_norm": 0.34328868985176086, | |
| "learning_rate": 9.682341441747446e-06, | |
| "loss": 0.1848, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 26.782608695652176, | |
| "grad_norm": 0.35909274220466614, | |
| "learning_rate": 9.665814072907293e-06, | |
| "loss": 0.437, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 26.956521739130434, | |
| "eval_loss": 0.47061336040496826, | |
| "eval_runtime": 11.1847, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 27.130434782608695, | |
| "grad_norm": 0.4798097312450409, | |
| "learning_rate": 9.648882429441258e-06, | |
| "loss": 0.3427, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 27.47826086956522, | |
| "grad_norm": 0.4749986529350281, | |
| "learning_rate": 9.63154797841033e-06, | |
| "loss": 0.3799, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 27.82608695652174, | |
| "grad_norm": 0.5031474828720093, | |
| "learning_rate": 9.613812221777212e-06, | |
| "loss": 0.3478, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 0.47599098086357117, | |
| "eval_runtime": 11.1847, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 28.17391304347826, | |
| "grad_norm": 0.38506221771240234, | |
| "learning_rate": 9.595676696276173e-06, | |
| "loss": 0.1651, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 28.52173913043478, | |
| "grad_norm": 0.6532711982727051, | |
| "learning_rate": 9.577142973279896e-06, | |
| "loss": 0.3973, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 28.869565217391305, | |
| "grad_norm": 0.4371902048587799, | |
| "learning_rate": 9.55821265866333e-06, | |
| "loss": 0.254, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 28.869565217391305, | |
| "eval_loss": 0.4744891822338104, | |
| "eval_runtime": 11.1866, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 29.217391304347824, | |
| "grad_norm": 0.4469977915287018, | |
| "learning_rate": 9.538887392664544e-06, | |
| "loss": 0.2313, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 29.565217391304348, | |
| "grad_norm": 0.42242884635925293, | |
| "learning_rate": 9.519168849742603e-06, | |
| "loss": 0.2858, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 29.91304347826087, | |
| "grad_norm": 0.47203147411346436, | |
| "learning_rate": 9.499058738432492e-06, | |
| "loss": 0.3934, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 29.91304347826087, | |
| "eval_loss": 0.48127391934394836, | |
| "eval_runtime": 11.1845, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 30.26086956521739, | |
| "grad_norm": 0.573071300983429, | |
| "learning_rate": 9.478558801197065e-06, | |
| "loss": 0.2361, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 30.608695652173914, | |
| "grad_norm": 0.6047350168228149, | |
| "learning_rate": 9.457670814276083e-06, | |
| "loss": 0.2041, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 30.956521739130434, | |
| "grad_norm": 0.5841593146324158, | |
| "learning_rate": 9.436396587532297e-06, | |
| "loss": 0.3074, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 30.956521739130434, | |
| "eval_loss": 0.4815356731414795, | |
| "eval_runtime": 11.1885, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 31.304347826086957, | |
| "grad_norm": 0.6498507857322693, | |
| "learning_rate": 9.414737964294636e-06, | |
| "loss": 0.3778, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 31.652173913043477, | |
| "grad_norm": 0.2765786945819855, | |
| "learning_rate": 9.392696821198488e-06, | |
| "loss": 0.1384, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "grad_norm": 0.5654636025428772, | |
| "learning_rate": 9.370275068023097e-06, | |
| "loss": 0.3447, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_loss": 0.482303649187088, | |
| "eval_runtime": 11.1841, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 32.34782608695652, | |
| "grad_norm": 0.4734094738960266, | |
| "learning_rate": 9.347474647526095e-06, | |
| "loss": 0.2441, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 32.69565217391305, | |
| "grad_norm": 0.5260018110275269, | |
| "learning_rate": 9.324297535275156e-06, | |
| "loss": 0.225, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 32.869565217391305, | |
| "eval_loss": 0.4865100383758545, | |
| "eval_runtime": 11.1875, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 32.869565217391305, | |
| "step": 189, | |
| "total_flos": 1.2027607575743693e+17, | |
| "train_loss": 0.9406116593767095, | |
| "train_runtime": 4446.8494, | |
| "train_samples_per_second": 1.552, | |
| "train_steps_per_second": 0.169 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 750, | |
| "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": 1.2027607575743693e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |