| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.7032967032967034, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.014065934065934066, | |
| "grad_norm": 0.37834057211875916, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3066, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.014065934065934066, | |
| "eval_loss": 1.3235194683074951, | |
| "eval_runtime": 41.1821, | |
| "eval_samples_per_second": 11.631, | |
| "eval_steps_per_second": 2.914, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.028131868131868132, | |
| "grad_norm": 0.3664795160293579, | |
| "learning_rate": 0.0001, | |
| "loss": 1.3373, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0421978021978022, | |
| "grad_norm": 0.576968789100647, | |
| "learning_rate": 9.989294616193017e-05, | |
| "loss": 1.322, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.056263736263736264, | |
| "grad_norm": 1.1225945949554443, | |
| "learning_rate": 9.957224306869053e-05, | |
| "loss": 1.2498, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.07032967032967033, | |
| "grad_norm": 0.5157985687255859, | |
| "learning_rate": 9.903926402016153e-05, | |
| "loss": 1.1756, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0843956043956044, | |
| "grad_norm": 0.7155728936195374, | |
| "learning_rate": 9.829629131445342e-05, | |
| "loss": 1.1287, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.09846153846153846, | |
| "grad_norm": 0.46939119696617126, | |
| "learning_rate": 9.73465064747553e-05, | |
| "loss": 1.0725, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.11252747252747253, | |
| "grad_norm": 0.3746448755264282, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 1.009, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.1265934065934066, | |
| "grad_norm": 0.3685707747936249, | |
| "learning_rate": 9.484363707663442e-05, | |
| "loss": 0.9849, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.14065934065934066, | |
| "grad_norm": 0.3576546907424927, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.9647, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15472527472527473, | |
| "grad_norm": 0.2630157470703125, | |
| "learning_rate": 9.157348061512727e-05, | |
| "loss": 0.9024, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.1687912087912088, | |
| "grad_norm": 0.22557401657104492, | |
| "learning_rate": 8.966766701456177e-05, | |
| "loss": 0.8968, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.18285714285714286, | |
| "grad_norm": 0.2797858715057373, | |
| "learning_rate": 8.759199037394887e-05, | |
| "loss": 0.8649, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.18285714285714286, | |
| "eval_loss": 0.8646656274795532, | |
| "eval_runtime": 41.2959, | |
| "eval_samples_per_second": 11.599, | |
| "eval_steps_per_second": 2.906, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.19692307692307692, | |
| "grad_norm": 0.2747080326080322, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.8738, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.210989010989011, | |
| "grad_norm": 0.2343476116657257, | |
| "learning_rate": 8.296729075500344e-05, | |
| "loss": 0.8745, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.22505494505494505, | |
| "grad_norm": 0.2237587720155716, | |
| "learning_rate": 8.043807145043604e-05, | |
| "loss": 0.8282, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.23912087912087912, | |
| "grad_norm": 0.19825835525989532, | |
| "learning_rate": 7.777851165098012e-05, | |
| "loss": 0.842, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.2531868131868132, | |
| "grad_norm": 0.16660410165786743, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.8453, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.2672527472527473, | |
| "grad_norm": 0.1577339917421341, | |
| "learning_rate": 7.211443451095007e-05, | |
| "loss": 0.844, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.2813186813186813, | |
| "grad_norm": 0.1596086323261261, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.8101, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2953846153846154, | |
| "grad_norm": 0.1533115953207016, | |
| "learning_rate": 6.607197326515808e-05, | |
| "loss": 0.8229, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.30945054945054945, | |
| "grad_norm": 0.1380361020565033, | |
| "learning_rate": 6.294095225512603e-05, | |
| "loss": 0.8142, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.32351648351648354, | |
| "grad_norm": 0.12135830521583557, | |
| "learning_rate": 5.9754516100806423e-05, | |
| "loss": 0.7935, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.3375824175824176, | |
| "grad_norm": 0.12721355259418488, | |
| "learning_rate": 5.6526309611002594e-05, | |
| "loss": 0.8114, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.3516483516483517, | |
| "grad_norm": 0.12719641625881195, | |
| "learning_rate": 5.327015646150716e-05, | |
| "loss": 0.8137, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.3657142857142857, | |
| "grad_norm": 0.12431051582098007, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8016, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.3657142857142857, | |
| "eval_loss": 0.7906247973442078, | |
| "eval_runtime": 41.3118, | |
| "eval_samples_per_second": 11.595, | |
| "eval_steps_per_second": 2.905, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.3797802197802198, | |
| "grad_norm": 0.1247800961136818, | |
| "learning_rate": 4.6729843538492847e-05, | |
| "loss": 0.7944, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.39384615384615385, | |
| "grad_norm": 0.11753375828266144, | |
| "learning_rate": 4.347369038899744e-05, | |
| "loss": 0.7916, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.40791208791208794, | |
| "grad_norm": 0.10586351901292801, | |
| "learning_rate": 4.0245483899193595e-05, | |
| "loss": 0.7824, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.421978021978022, | |
| "grad_norm": 0.10135481506586075, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 0.7672, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.43604395604395607, | |
| "grad_norm": 0.10301663726568222, | |
| "learning_rate": 3.392802673484193e-05, | |
| "loss": 0.7961, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.4501098901098901, | |
| "grad_norm": 0.1041460931301117, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.7891, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.4641758241758242, | |
| "grad_norm": 0.09919846802949905, | |
| "learning_rate": 2.7885565489049946e-05, | |
| "loss": 0.8119, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.47824175824175824, | |
| "grad_norm": 0.09898246824741364, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.7787, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.49230769230769234, | |
| "grad_norm": 0.10475801676511765, | |
| "learning_rate": 2.2221488349019903e-05, | |
| "loss": 0.7885, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.5063736263736264, | |
| "grad_norm": 0.09680575132369995, | |
| "learning_rate": 1.9561928549563968e-05, | |
| "loss": 0.7795, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.5204395604395604, | |
| "grad_norm": 0.10325780510902405, | |
| "learning_rate": 1.703270924499656e-05, | |
| "loss": 0.7897, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.5345054945054946, | |
| "grad_norm": 0.09474309533834457, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.7936, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.5485714285714286, | |
| "grad_norm": 0.09674929827451706, | |
| "learning_rate": 1.2408009626051137e-05, | |
| "loss": 0.8068, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.5485714285714286, | |
| "eval_loss": 0.7735550403594971, | |
| "eval_runtime": 41.3006, | |
| "eval_samples_per_second": 11.598, | |
| "eval_steps_per_second": 2.906, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.5626373626373626, | |
| "grad_norm": 0.09625887870788574, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 0.7666, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5767032967032967, | |
| "grad_norm": 0.09051655977964401, | |
| "learning_rate": 8.426519384872733e-06, | |
| "loss": 0.7939, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.5907692307692308, | |
| "grad_norm": 0.09801479429006577, | |
| "learning_rate": 6.698729810778065e-06, | |
| "loss": 0.762, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.6048351648351649, | |
| "grad_norm": 0.0950697809457779, | |
| "learning_rate": 5.156362923365588e-06, | |
| "loss": 0.8127, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.6189010989010989, | |
| "grad_norm": 0.08845943212509155, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.7691, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.6329670329670329, | |
| "grad_norm": 0.09417203068733215, | |
| "learning_rate": 2.653493525244721e-06, | |
| "loss": 0.7847, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.6470329670329671, | |
| "grad_norm": 0.09285993129014969, | |
| "learning_rate": 1.70370868554659e-06, | |
| "loss": 0.7798, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.6610989010989011, | |
| "grad_norm": 0.09379366785287857, | |
| "learning_rate": 9.607359798384785e-07, | |
| "loss": 0.7899, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.6751648351648352, | |
| "grad_norm": 0.08996466547250748, | |
| "learning_rate": 4.277569313094809e-07, | |
| "loss": 0.8066, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.6892307692307692, | |
| "grad_norm": 0.09068991243839264, | |
| "learning_rate": 1.0705383806982606e-07, | |
| "loss": 0.7763, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.7032967032967034, | |
| "grad_norm": 0.089011549949646, | |
| "learning_rate": 0.0, | |
| "loss": 0.7753, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0547690207182848e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |