| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.43383947939262474, | |
| "eval_steps": 25, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005784526391901663, | |
| "grad_norm": 14.124544143676758, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 20.3452, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.005784526391901663, | |
| "eval_loss": NaN, | |
| "eval_runtime": 16.2429, | |
| "eval_samples_per_second": 8.989, | |
| "eval_steps_per_second": 4.494, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.011569052783803326, | |
| "grad_norm": 12.12727165222168, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 21.4742, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.01735357917570499, | |
| "grad_norm": 12.951985359191895, | |
| "learning_rate": 0.0001, | |
| "loss": 20.3039, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.023138105567606652, | |
| "grad_norm": 17.767934799194336, | |
| "learning_rate": 9.99524110790929e-05, | |
| "loss": 19.245, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.028922631959508314, | |
| "grad_norm": 21.18607521057129, | |
| "learning_rate": 9.980973490458728e-05, | |
| "loss": 16.6112, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.03470715835140998, | |
| "grad_norm": 32.60835266113281, | |
| "learning_rate": 9.957224306869053e-05, | |
| "loss": 12.6665, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.04049168474331164, | |
| "grad_norm": 25.802278518676758, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 9.0528, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.046276211135213303, | |
| "grad_norm": 16.02645492553711, | |
| "learning_rate": 9.881480035599667e-05, | |
| "loss": 7.16, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.052060737527114966, | |
| "grad_norm": 13.585994720458984, | |
| "learning_rate": 9.829629131445342e-05, | |
| "loss": 6.3665, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.05784526391901663, | |
| "grad_norm": 14.590239524841309, | |
| "learning_rate": 9.768584753741134e-05, | |
| "loss": 5.3905, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0636297903109183, | |
| "grad_norm": 6.89243221282959, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 4.801, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.06941431670281996, | |
| "grad_norm": 10.386966705322266, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 4.4869, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.07519884309472162, | |
| "grad_norm": 11.136741638183594, | |
| "learning_rate": 9.53153893518325e-05, | |
| "loss": 4.7254, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.08098336948662328, | |
| "grad_norm": 16.662979125976562, | |
| "learning_rate": 9.435054165891109e-05, | |
| "loss": 4.6043, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.08676789587852494, | |
| "grad_norm": 4.838995933532715, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 4.1652, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.09255242227042661, | |
| "grad_norm": 6.366863250732422, | |
| "learning_rate": 9.21695722906443e-05, | |
| "loss": 4.0379, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.09833694866232827, | |
| "grad_norm": 9.666878700256348, | |
| "learning_rate": 9.09576022144496e-05, | |
| "loss": 3.853, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.10412147505422993, | |
| "grad_norm": 8.335044860839844, | |
| "learning_rate": 8.966766701456177e-05, | |
| "loss": 3.8593, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.1099060014461316, | |
| "grad_norm": 5.919156551361084, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 3.7728, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.11569052783803326, | |
| "grad_norm": 2.7403481006622314, | |
| "learning_rate": 8.68638668405062e-05, | |
| "loss": 3.5911, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.12147505422993492, | |
| "grad_norm": 2.4810056686401367, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 3.493, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.1272595806218366, | |
| "grad_norm": 2.9197349548339844, | |
| "learning_rate": 8.377951038078302e-05, | |
| "loss": 3.4103, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.13304410701373826, | |
| "grad_norm": 7.832423686981201, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 3.5971, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.13882863340563992, | |
| "grad_norm": 10.133871078491211, | |
| "learning_rate": 8.043807145043604e-05, | |
| "loss": 3.8441, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.14461315979754158, | |
| "grad_norm": 3.711956262588501, | |
| "learning_rate": 7.86788218175523e-05, | |
| "loss": 3.3329, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.14461315979754158, | |
| "eval_loss": NaN, | |
| "eval_runtime": 16.2967, | |
| "eval_samples_per_second": 8.959, | |
| "eval_steps_per_second": 4.479, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.15039768618944324, | |
| "grad_norm": 5.868927001953125, | |
| "learning_rate": 7.68649804173412e-05, | |
| "loss": 3.1901, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.1561822125813449, | |
| "grad_norm": 2.6448488235473633, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 3.1779, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.16196673897324657, | |
| "grad_norm": 1.8636095523834229, | |
| "learning_rate": 7.308743066175172e-05, | |
| "loss": 3.1947, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.16775126536514823, | |
| "grad_norm": 4.58578634262085, | |
| "learning_rate": 7.113091308703498e-05, | |
| "loss": 3.5627, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.1735357917570499, | |
| "grad_norm": 21.534526824951172, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 3.68, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.17932031814895155, | |
| "grad_norm": 8.905357360839844, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 3.7521, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.18510484454085321, | |
| "grad_norm": 4.796418190002441, | |
| "learning_rate": 6.503528997521366e-05, | |
| "loss": 3.2734, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.19088937093275488, | |
| "grad_norm": 2.0651814937591553, | |
| "learning_rate": 6.294095225512603e-05, | |
| "loss": 3.0878, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.19667389732465654, | |
| "grad_norm": 3.3176019191741943, | |
| "learning_rate": 6.0821980696905146e-05, | |
| "loss": 3.186, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.2024584237165582, | |
| "grad_norm": 2.418898344039917, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 3.0962, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.20824295010845986, | |
| "grad_norm": 4.730885028839111, | |
| "learning_rate": 5.6526309611002594e-05, | |
| "loss": 3.1704, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.21402747650036152, | |
| "grad_norm": 4.24273681640625, | |
| "learning_rate": 5.435778713738292e-05, | |
| "loss": 3.2574, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.2198120028922632, | |
| "grad_norm": 9.04359245300293, | |
| "learning_rate": 5.218096936826681e-05, | |
| "loss": 3.4872, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.22559652928416485, | |
| "grad_norm": 0.8422988653182983, | |
| "learning_rate": 5e-05, | |
| "loss": 3.0299, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.2313810556760665, | |
| "grad_norm": 8.053430557250977, | |
| "learning_rate": 4.781903063173321e-05, | |
| "loss": 3.4567, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.23716558206796817, | |
| "grad_norm": 0.7461191415786743, | |
| "learning_rate": 4.564221286261709e-05, | |
| "loss": 3.0264, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.24295010845986983, | |
| "grad_norm": 0.6736895442008972, | |
| "learning_rate": 4.347369038899744e-05, | |
| "loss": 3.0185, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.24873463485177152, | |
| "grad_norm": 0.15485769510269165, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 3.0018, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.2545191612436732, | |
| "grad_norm": 5.6190338134765625, | |
| "learning_rate": 3.917801930309486e-05, | |
| "loss": 3.7042, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.2603036876355748, | |
| "grad_norm": 17.187482833862305, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 4.8073, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2660882140274765, | |
| "grad_norm": 2.7848260402679443, | |
| "learning_rate": 3.4964710024786354e-05, | |
| "loss": 3.3303, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.27187274041937814, | |
| "grad_norm": 7.332921504974365, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 3.6265, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.27765726681127983, | |
| "grad_norm": 6.031116962432861, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 3.26, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.28344179320318147, | |
| "grad_norm": 5.31016206741333, | |
| "learning_rate": 2.886908691296504e-05, | |
| "loss": 3.4005, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.28922631959508316, | |
| "grad_norm": 6.793763160705566, | |
| "learning_rate": 2.6912569338248315e-05, | |
| "loss": 3.6388, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.28922631959508316, | |
| "eval_loss": NaN, | |
| "eval_runtime": 16.2685, | |
| "eval_samples_per_second": 8.974, | |
| "eval_steps_per_second": 4.487, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2950108459869848, | |
| "grad_norm": 3.044539213180542, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 3.491, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.3007953723788865, | |
| "grad_norm": 8.76490592956543, | |
| "learning_rate": 2.3135019582658802e-05, | |
| "loss": 3.5022, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.3065798987707881, | |
| "grad_norm": 11.707574844360352, | |
| "learning_rate": 2.132117818244771e-05, | |
| "loss": 3.6092, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.3123644251626898, | |
| "grad_norm": 12.732514381408691, | |
| "learning_rate": 1.9561928549563968e-05, | |
| "loss": 3.4725, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.31814895155459144, | |
| "grad_norm": 7.573165416717529, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 3.3587, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.32393347794649313, | |
| "grad_norm": 5.776604652404785, | |
| "learning_rate": 1.622048961921699e-05, | |
| "loss": 3.2847, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.3297180043383948, | |
| "grad_norm": 9.322304725646973, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 3.4005, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.33550253073029646, | |
| "grad_norm": 1.8316805362701416, | |
| "learning_rate": 1.3136133159493802e-05, | |
| "loss": 3.2024, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.34128705712219815, | |
| "grad_norm": 1.2074944972991943, | |
| "learning_rate": 1.1697777844051105e-05, | |
| "loss": 3.1161, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.3470715835140998, | |
| "grad_norm": 6.147497653961182, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 3.7194, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.35285610990600147, | |
| "grad_norm": 5.360683917999268, | |
| "learning_rate": 9.042397785550405e-06, | |
| "loss": 3.3806, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.3586406362979031, | |
| "grad_norm": 3.081329345703125, | |
| "learning_rate": 7.830427709355725e-06, | |
| "loss": 3.1167, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.3644251626898048, | |
| "grad_norm": 6.260736465454102, | |
| "learning_rate": 6.698729810778065e-06, | |
| "loss": 3.4621, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.37020968908170643, | |
| "grad_norm": 1.834079623222351, | |
| "learning_rate": 5.649458341088915e-06, | |
| "loss": 3.1454, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.3759942154736081, | |
| "grad_norm": 3.3285751342773438, | |
| "learning_rate": 4.684610648167503e-06, | |
| "loss": 3.2656, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.38177874186550975, | |
| "grad_norm": 6.730548858642578, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 3.4244, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.38756326825741144, | |
| "grad_norm": 2.947965145111084, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 3.161, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.3933477946493131, | |
| "grad_norm": 5.567538261413574, | |
| "learning_rate": 2.314152462588659e-06, | |
| "loss": 3.3722, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.39913232104121477, | |
| "grad_norm": 4.872069358825684, | |
| "learning_rate": 1.70370868554659e-06, | |
| "loss": 3.388, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.4049168474331164, | |
| "grad_norm": 0.6829429864883423, | |
| "learning_rate": 1.1851996440033319e-06, | |
| "loss": 3.0004, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.4107013738250181, | |
| "grad_norm": 2.973910093307495, | |
| "learning_rate": 7.596123493895991e-07, | |
| "loss": 3.2397, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.4164859002169197, | |
| "grad_norm": 2.918032646179199, | |
| "learning_rate": 4.277569313094809e-07, | |
| "loss": 3.1185, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.4222704266088214, | |
| "grad_norm": 3.0201897621154785, | |
| "learning_rate": 1.9026509541272275e-07, | |
| "loss": 3.418, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.42805495300072305, | |
| "grad_norm": 2.436922550201416, | |
| "learning_rate": 4.7588920907110094e-08, | |
| "loss": 3.2234, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.43383947939262474, | |
| "grad_norm": 5.052024841308594, | |
| "learning_rate": 0.0, | |
| "loss": 3.3554, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.43383947939262474, | |
| "eval_loss": NaN, | |
| "eval_runtime": 16.2973, | |
| "eval_samples_per_second": 8.959, | |
| "eval_steps_per_second": 4.479, | |
| "step": 75 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 75, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.119015678246912e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |