| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 59.0, | |
| "eval_steps": 500, | |
| "global_step": 177, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.6283529996871948, | |
| "eval_runtime": 3.1863, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.4084081649780273, | |
| "eval_runtime": 3.1869, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.2192631959915161, | |
| "eval_runtime": 3.1877, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 0.43165192008018494, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4843, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.040535569190979, | |
| "eval_runtime": 3.1877, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.8558390736579895, | |
| "eval_runtime": 3.187, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.6931569576263428, | |
| "eval_runtime": 3.1865, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 0.8776572942733765, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7175, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.5830169916152954, | |
| "eval_runtime": 3.1867, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.49080416560173035, | |
| "eval_runtime": 3.1854, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.48263129591941833, | |
| "eval_runtime": 3.1874, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.8698680996894836, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1599, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.470937579870224, | |
| "eval_runtime": 3.1861, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.4739013612270355, | |
| "eval_runtime": 3.1862, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.4649692475795746, | |
| "eval_runtime": 3.1861, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.4578638970851898, | |
| "eval_runtime": 3.1863, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 13.333333333333334, | |
| "grad_norm": 0.8457691669464111, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0259, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.4607035219669342, | |
| "eval_runtime": 3.186, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.45176398754119873, | |
| "eval_runtime": 3.1856, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.4511236846446991, | |
| "eval_runtime": 3.1849, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 16.666666666666668, | |
| "grad_norm": 0.5995622277259827, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0099, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 0.43412062525749207, | |
| "eval_runtime": 3.1865, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.43398618698120117, | |
| "eval_runtime": 3.1863, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 0.43465808033943176, | |
| "eval_runtime": 3.1849, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.09180915355682373, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0031, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.43288353085517883, | |
| "eval_runtime": 3.187, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_loss": 0.43253418803215027, | |
| "eval_runtime": 3.186, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 0.44188740849494934, | |
| "eval_runtime": 3.1856, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_loss": 0.43514636158943176, | |
| "eval_runtime": 3.1859, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 23.333333333333332, | |
| "grad_norm": 0.010112113319337368, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0031, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 0.43539562821388245, | |
| "eval_runtime": 3.1858, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 0.4404897689819336, | |
| "eval_runtime": 3.186, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 0.4413206875324249, | |
| "eval_runtime": 3.1867, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 26.666666666666668, | |
| "grad_norm": 0.006296033971011639, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0021, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_loss": 0.4521159827709198, | |
| "eval_runtime": 3.1866, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 0.44687846302986145, | |
| "eval_runtime": 3.1854, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_loss": 0.4467637240886688, | |
| "eval_runtime": 3.1856, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 0.33079442381858826, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0029, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 0.44668495655059814, | |
| "eval_runtime": 3.1874, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_loss": 0.4493657052516937, | |
| "eval_runtime": 3.1855, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_loss": 0.45611047744750977, | |
| "eval_runtime": 3.1877, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_loss": 0.45603322982788086, | |
| "eval_runtime": 3.1856, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 33.333333333333336, | |
| "grad_norm": 0.0036402323748916388, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0013, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_loss": 0.4647960662841797, | |
| "eval_runtime": 3.1861, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_loss": 0.471958190202713, | |
| "eval_runtime": 3.1864, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_loss": 0.46781960129737854, | |
| "eval_runtime": 3.1871, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 36.666666666666664, | |
| "grad_norm": 0.0058022052980959415, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0005, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_loss": 0.46317848563194275, | |
| "eval_runtime": 3.1866, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_loss": 0.4615744352340698, | |
| "eval_runtime": 3.1856, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_loss": 0.4577348232269287, | |
| "eval_runtime": 3.187, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 0.0034027991350740194, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0002, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 0.45408549904823303, | |
| "eval_runtime": 3.1867, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_loss": 0.4540380537509918, | |
| "eval_runtime": 3.1872, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_loss": 0.45183610916137695, | |
| "eval_runtime": 3.1862, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_loss": 0.4519132077693939, | |
| "eval_runtime": 3.1863, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 43.333333333333336, | |
| "grad_norm": 0.0013499228516593575, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0001, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_loss": 0.45116329193115234, | |
| "eval_runtime": 3.1866, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_loss": 0.4517379105091095, | |
| "eval_runtime": 3.187, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_loss": 0.4512501657009125, | |
| "eval_runtime": 3.1865, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 46.666666666666664, | |
| "grad_norm": 0.0008508643950335681, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0001, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_loss": 0.45101478695869446, | |
| "eval_runtime": 3.1863, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.942, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_loss": 0.45259496569633484, | |
| "eval_runtime": 3.1879, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_loss": 0.4528372287750244, | |
| "eval_runtime": 3.1871, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 0.0005910882027819753, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0001, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_loss": 0.45282843708992004, | |
| "eval_runtime": 3.1901, | |
| "eval_samples_per_second": 0.94, | |
| "eval_steps_per_second": 0.94, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_loss": 0.450717955827713, | |
| "eval_runtime": 3.1877, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_loss": 0.45283257961273193, | |
| "eval_runtime": 3.1886, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_loss": 0.45390304923057556, | |
| "eval_runtime": 3.1866, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 53.333333333333336, | |
| "grad_norm": 0.0004977145581506193, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_loss": 0.45244312286376953, | |
| "eval_runtime": 3.1865, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_loss": 0.45304372906684875, | |
| "eval_runtime": 3.1878, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_loss": 0.4535575211048126, | |
| "eval_runtime": 3.187, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 56.666666666666664, | |
| "grad_norm": 0.0004469017731025815, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_loss": 0.45485830307006836, | |
| "eval_runtime": 3.1882, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_loss": 0.45483624935150146, | |
| "eval_runtime": 3.1869, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_loss": 0.453870564699173, | |
| "eval_runtime": 3.1879, | |
| "eval_samples_per_second": 0.941, | |
| "eval_steps_per_second": 0.941, | |
| "step": 177 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 180, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 60, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.822242672685875e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |