| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.36977120406748326, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.007395424081349665, | |
| "grad_norm": 507.5455017089844, | |
| "learning_rate": 5e-05, | |
| "loss": 119.865, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.007395424081349665, | |
| "eval_loss": 3.9573512077331543, | |
| "eval_runtime": 81.2876, | |
| "eval_samples_per_second": 11.207, | |
| "eval_steps_per_second": 2.805, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01479084816269933, | |
| "grad_norm": 472.2460632324219, | |
| "learning_rate": 0.0001, | |
| "loss": 123.5993, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.022186272244048996, | |
| "grad_norm": 214.4851531982422, | |
| "learning_rate": 9.989294616193017e-05, | |
| "loss": 120.4008, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.02958169632539866, | |
| "grad_norm": 170.87863159179688, | |
| "learning_rate": 9.957224306869053e-05, | |
| "loss": 114.044, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.036977120406748325, | |
| "grad_norm": 318.2630920410156, | |
| "learning_rate": 9.903926402016153e-05, | |
| "loss": 109.6968, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.04437254448809799, | |
| "grad_norm": 166.34352111816406, | |
| "learning_rate": 9.829629131445342e-05, | |
| "loss": 105.4008, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.05176796856944765, | |
| "grad_norm": 97.02327728271484, | |
| "learning_rate": 9.73465064747553e-05, | |
| "loss": 95.6925, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.05916339265079732, | |
| "grad_norm": 103.01248931884766, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 92.4355, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.06655881673214699, | |
| "grad_norm": 115.37718200683594, | |
| "learning_rate": 9.484363707663442e-05, | |
| "loss": 97.7696, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.07395424081349665, | |
| "grad_norm": 74.1399917602539, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 88.485, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08134966489484631, | |
| "grad_norm": 107.24742126464844, | |
| "learning_rate": 9.157348061512727e-05, | |
| "loss": 90.4529, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.08874508897619598, | |
| "grad_norm": 102.1383285522461, | |
| "learning_rate": 8.966766701456177e-05, | |
| "loss": 88.3525, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.09614051305754565, | |
| "grad_norm": 57.68673324584961, | |
| "learning_rate": 8.759199037394887e-05, | |
| "loss": 85.8088, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.09614051305754565, | |
| "eval_loss": 2.8494491577148438, | |
| "eval_runtime": 81.3898, | |
| "eval_samples_per_second": 11.193, | |
| "eval_steps_per_second": 2.801, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.1035359371388953, | |
| "grad_norm": 69.56649017333984, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 86.8469, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.11093136122024497, | |
| "grad_norm": 67.40845489501953, | |
| "learning_rate": 8.296729075500344e-05, | |
| "loss": 88.4199, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.11832678530159464, | |
| "grad_norm": 70.2257308959961, | |
| "learning_rate": 8.043807145043604e-05, | |
| "loss": 91.2391, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.12572220938294432, | |
| "grad_norm": 46.68100357055664, | |
| "learning_rate": 7.777851165098012e-05, | |
| "loss": 87.3729, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.13311763346429398, | |
| "grad_norm": 102.54607391357422, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 82.5651, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.14051305754564364, | |
| "grad_norm": 116.23051452636719, | |
| "learning_rate": 7.211443451095007e-05, | |
| "loss": 93.5197, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.1479084816269933, | |
| "grad_norm": 65.93101501464844, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 82.7672, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.15530390570834296, | |
| "grad_norm": 57.98164749145508, | |
| "learning_rate": 6.607197326515808e-05, | |
| "loss": 85.3329, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.16269932978969262, | |
| "grad_norm": 60.674495697021484, | |
| "learning_rate": 6.294095225512603e-05, | |
| "loss": 84.3847, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.17009475387104228, | |
| "grad_norm": 72.81175231933594, | |
| "learning_rate": 5.9754516100806423e-05, | |
| "loss": 88.0836, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.17749017795239197, | |
| "grad_norm": 57.464141845703125, | |
| "learning_rate": 5.6526309611002594e-05, | |
| "loss": 83.9486, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.18488560203374163, | |
| "grad_norm": 41.056941986083984, | |
| "learning_rate": 5.327015646150716e-05, | |
| "loss": 79.0182, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.1922810261150913, | |
| "grad_norm": 55.79506301879883, | |
| "learning_rate": 5e-05, | |
| "loss": 85.1922, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.1922810261150913, | |
| "eval_loss": 2.7002317905426025, | |
| "eval_runtime": 81.3945, | |
| "eval_samples_per_second": 11.192, | |
| "eval_steps_per_second": 2.801, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.19967645019644095, | |
| "grad_norm": 56.23450469970703, | |
| "learning_rate": 4.6729843538492847e-05, | |
| "loss": 85.5899, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.2070718742777906, | |
| "grad_norm": 55.729068756103516, | |
| "learning_rate": 4.347369038899744e-05, | |
| "loss": 84.6372, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.21446729835914027, | |
| "grad_norm": 44.94795608520508, | |
| "learning_rate": 4.0245483899193595e-05, | |
| "loss": 81.2659, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.22186272244048993, | |
| "grad_norm": 43.78959274291992, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 84.2274, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.22925814652183962, | |
| "grad_norm": 40.38026809692383, | |
| "learning_rate": 3.392802673484193e-05, | |
| "loss": 80.1799, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.23665357060318928, | |
| "grad_norm": 39.87312698364258, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 80.2665, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.24404899468453894, | |
| "grad_norm": 42.08469772338867, | |
| "learning_rate": 2.7885565489049946e-05, | |
| "loss": 81.3834, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.25144441876588863, | |
| "grad_norm": 40.6140251159668, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 80.8223, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.25883984284723827, | |
| "grad_norm": 52.44778823852539, | |
| "learning_rate": 2.2221488349019903e-05, | |
| "loss": 85.2504, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.26623526692858795, | |
| "grad_norm": 44.90240478515625, | |
| "learning_rate": 1.9561928549563968e-05, | |
| "loss": 80.1232, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.2736306910099376, | |
| "grad_norm": 44.05722427368164, | |
| "learning_rate": 1.703270924499656e-05, | |
| "loss": 85.218, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.2810261150912873, | |
| "grad_norm": 34.475032806396484, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 74.1421, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.2884215391726369, | |
| "grad_norm": 50.83314514160156, | |
| "learning_rate": 1.2408009626051137e-05, | |
| "loss": 85.2003, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.2884215391726369, | |
| "eval_loss": 2.658181667327881, | |
| "eval_runtime": 81.3831, | |
| "eval_samples_per_second": 11.194, | |
| "eval_steps_per_second": 2.802, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.2958169632539866, | |
| "grad_norm": 40.7790412902832, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 80.1685, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3032123873353363, | |
| "grad_norm": 44.15685272216797, | |
| "learning_rate": 8.426519384872733e-06, | |
| "loss": 83.519, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.3106078114166859, | |
| "grad_norm": 33.223167419433594, | |
| "learning_rate": 6.698729810778065e-06, | |
| "loss": 74.8312, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.3180032354980356, | |
| "grad_norm": 42.13229751586914, | |
| "learning_rate": 5.156362923365588e-06, | |
| "loss": 79.9294, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.32539865957938524, | |
| "grad_norm": 36.99294662475586, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 78.9502, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.33279408366073493, | |
| "grad_norm": 40.4399528503418, | |
| "learning_rate": 2.653493525244721e-06, | |
| "loss": 82.6471, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.34018950774208456, | |
| "grad_norm": 40.24978256225586, | |
| "learning_rate": 1.70370868554659e-06, | |
| "loss": 82.7816, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.34758493182343425, | |
| "grad_norm": 42.67814254760742, | |
| "learning_rate": 9.607359798384785e-07, | |
| "loss": 80.2397, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.35498035590478394, | |
| "grad_norm": 37.027217864990234, | |
| "learning_rate": 4.277569313094809e-07, | |
| "loss": 80.1654, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.3623757799861336, | |
| "grad_norm": 44.58671951293945, | |
| "learning_rate": 1.0705383806982606e-07, | |
| "loss": 79.761, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.36977120406748326, | |
| "grad_norm": 44.38752746582031, | |
| "learning_rate": 0.0, | |
| "loss": 76.1375, | |
| "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": 1.1250084863803392e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |