| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.05841974587410545, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0011683949174821088, | |
| "grad_norm": 0.5629591941833496, | |
| "learning_rate": 2e-05, | |
| "loss": 1.6259, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0011683949174821088, | |
| "eval_loss": 1.7091846466064453, | |
| "eval_runtime": 32.7497, | |
| "eval_samples_per_second": 22.016, | |
| "eval_steps_per_second": 11.023, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0023367898349642177, | |
| "grad_norm": 0.4464726150035858, | |
| "learning_rate": 4e-05, | |
| "loss": 1.9928, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0035051847524463268, | |
| "grad_norm": 0.554092288017273, | |
| "learning_rate": 6e-05, | |
| "loss": 1.9481, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.004673579669928435, | |
| "grad_norm": 0.5418124198913574, | |
| "learning_rate": 8e-05, | |
| "loss": 2.0157, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0058419745874105445, | |
| "grad_norm": 0.6429816484451294, | |
| "learning_rate": 0.0001, | |
| "loss": 1.9883, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0058419745874105445, | |
| "eval_loss": 1.6724804639816284, | |
| "eval_runtime": 31.1462, | |
| "eval_samples_per_second": 23.149, | |
| "eval_steps_per_second": 11.59, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0070103695048926535, | |
| "grad_norm": 0.7108197212219238, | |
| "learning_rate": 0.00012, | |
| "loss": 1.9129, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.008178764422374763, | |
| "grad_norm": 0.6660429835319519, | |
| "learning_rate": 0.00014, | |
| "loss": 1.5668, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.00934715933985687, | |
| "grad_norm": 0.7972020506858826, | |
| "learning_rate": 0.00016, | |
| "loss": 1.6182, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.01051555425733898, | |
| "grad_norm": 0.9052525162696838, | |
| "learning_rate": 0.00018, | |
| "loss": 1.7629, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.011683949174821089, | |
| "grad_norm": 0.6085521578788757, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7345, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.011683949174821089, | |
| "eval_loss": 1.4248594045639038, | |
| "eval_runtime": 31.1654, | |
| "eval_samples_per_second": 23.135, | |
| "eval_steps_per_second": 11.583, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.012852344092303198, | |
| "grad_norm": 0.6540700793266296, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 1.3627, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.014020739009785307, | |
| "grad_norm": 1.1813312768936157, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 1.4923, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.015189133927267416, | |
| "grad_norm": 0.9703741669654846, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 1.5092, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.016357528844749527, | |
| "grad_norm": 1.16099214553833, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 1.2979, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.017525923762231634, | |
| "grad_norm": 1.7793681621551514, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 1.8037, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.017525923762231634, | |
| "eval_loss": 1.2205755710601807, | |
| "eval_runtime": 31.1965, | |
| "eval_samples_per_second": 23.112, | |
| "eval_steps_per_second": 11.572, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.01869431867971374, | |
| "grad_norm": 1.3740328550338745, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 1.2364, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.019862713597195852, | |
| "grad_norm": 0.6489018201828003, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 1.7165, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.02103110851467796, | |
| "grad_norm": 1.0925291776657104, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 1.4376, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.02219950343216007, | |
| "grad_norm": 0.28496918082237244, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 1.5726, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.023367898349642178, | |
| "grad_norm": 0.37671753764152527, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 1.1192, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.023367898349642178, | |
| "eval_loss": 1.2007029056549072, | |
| "eval_runtime": 31.1968, | |
| "eval_samples_per_second": 23.111, | |
| "eval_steps_per_second": 11.572, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02453629326712429, | |
| "grad_norm": 0.3938184380531311, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 1.4101, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.025704688184606396, | |
| "grad_norm": 0.42164233326911926, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 1.1861, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.026873083102088507, | |
| "grad_norm": 0.3876625597476959, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 1.3193, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.028041478019570614, | |
| "grad_norm": 0.511690080165863, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 1.1979, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.029209872937052725, | |
| "grad_norm": 0.4257306456565857, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 1.4168, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.029209872937052725, | |
| "eval_loss": 1.1946120262145996, | |
| "eval_runtime": 31.1806, | |
| "eval_samples_per_second": 23.123, | |
| "eval_steps_per_second": 11.578, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.030378267854534832, | |
| "grad_norm": 0.3946686387062073, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 1.3165, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.03154666277201694, | |
| "grad_norm": 0.37801477313041687, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 1.6097, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.032715057689499054, | |
| "grad_norm": 0.3698787987232208, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 1.4314, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.03388345260698116, | |
| "grad_norm": 0.36553719639778137, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 1.6258, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.03505184752446327, | |
| "grad_norm": 0.3180171847343445, | |
| "learning_rate": 0.0001, | |
| "loss": 1.8529, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03505184752446327, | |
| "eval_loss": 1.188832402229309, | |
| "eval_runtime": 31.1804, | |
| "eval_samples_per_second": 23.123, | |
| "eval_steps_per_second": 11.578, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.036220242441945376, | |
| "grad_norm": 0.3740488886833191, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 0.7608, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.03738863735942748, | |
| "grad_norm": 0.3934629261493683, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 1.1501, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.0385570322769096, | |
| "grad_norm": 0.3714420199394226, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 1.2233, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.039725427194391705, | |
| "grad_norm": 0.3141462206840515, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 1.3308, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.04089382211187381, | |
| "grad_norm": 0.3738103210926056, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 1.5121, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.04089382211187381, | |
| "eval_loss": 1.1856106519699097, | |
| "eval_runtime": 31.1551, | |
| "eval_samples_per_second": 23.142, | |
| "eval_steps_per_second": 11.587, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.04206221702935592, | |
| "grad_norm": 0.3965364992618561, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 1.6638, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.043230611946838034, | |
| "grad_norm": 0.40209704637527466, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 1.2929, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.04439900686432014, | |
| "grad_norm": 0.4041275084018707, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 1.3061, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.04556740178180225, | |
| "grad_norm": 0.3122788965702057, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 1.3712, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.046735796699284356, | |
| "grad_norm": 0.2880614697933197, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 1.7083, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.046735796699284356, | |
| "eval_loss": 1.18394935131073, | |
| "eval_runtime": 31.17, | |
| "eval_samples_per_second": 23.131, | |
| "eval_steps_per_second": 11.582, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.04790419161676647, | |
| "grad_norm": 0.4715648293495178, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 1.2266, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.04907258653424858, | |
| "grad_norm": 0.3626469075679779, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 1.3755, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.050240981451730685, | |
| "grad_norm": 0.3921823799610138, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 1.3125, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.05140937636921279, | |
| "grad_norm": 0.35432127118110657, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 1.2351, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.052577771286694906, | |
| "grad_norm": 0.4417038559913635, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 1.7417, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.052577771286694906, | |
| "eval_loss": 1.1834783554077148, | |
| "eval_runtime": 31.2195, | |
| "eval_samples_per_second": 23.095, | |
| "eval_steps_per_second": 11.563, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.053746166204177014, | |
| "grad_norm": 0.3598865866661072, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 1.4514, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.05491456112165912, | |
| "grad_norm": 0.43709179759025574, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 1.5828, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.05608295603914123, | |
| "grad_norm": 0.32742375135421753, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 1.5362, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.057251350956623336, | |
| "grad_norm": 0.31574079394340515, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 1.5458, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.05841974587410545, | |
| "grad_norm": 0.3892696797847748, | |
| "learning_rate": 0.0, | |
| "loss": 1.5221, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.05841974587410545, | |
| "eval_loss": 1.1832040548324585, | |
| "eval_runtime": 31.1605, | |
| "eval_samples_per_second": 23.138, | |
| "eval_steps_per_second": 11.585, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.3069219818110976e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |