| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.4767580452920143, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004767580452920143, | |
| "eval_loss": 3.438885450363159, | |
| "eval_runtime": 23.2999, | |
| "eval_samples_per_second": 15.193, | |
| "eval_steps_per_second": 1.931, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.014302741358760428, | |
| "grad_norm": 1.0672996044158936, | |
| "learning_rate": 3e-05, | |
| "loss": 3.2121, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.028605482717520857, | |
| "grad_norm": 0.7437488436698914, | |
| "learning_rate": 6e-05, | |
| "loss": 3.1757, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.04290822407628129, | |
| "grad_norm": 1.009957194328308, | |
| "learning_rate": 9e-05, | |
| "loss": 3.1921, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.04290822407628129, | |
| "eval_loss": 3.1295535564422607, | |
| "eval_runtime": 23.6302, | |
| "eval_samples_per_second": 14.981, | |
| "eval_steps_per_second": 1.904, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.057210965435041714, | |
| "grad_norm": 0.9846349358558655, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 3.0455, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.07151370679380215, | |
| "grad_norm": 1.2678148746490479, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 2.9, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.08581644815256258, | |
| "grad_norm": 0.8464758396148682, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 2.647, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.08581644815256258, | |
| "eval_loss": 2.543804168701172, | |
| "eval_runtime": 23.7134, | |
| "eval_samples_per_second": 14.928, | |
| "eval_steps_per_second": 1.898, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.10011918951132301, | |
| "grad_norm": 0.8965421915054321, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 2.4721, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.11442193087008343, | |
| "grad_norm": 0.8901471495628357, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 2.4806, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.12872467222884387, | |
| "grad_norm": 0.6464989185333252, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 2.4984, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.12872467222884387, | |
| "eval_loss": 2.3897674083709717, | |
| "eval_runtime": 23.7248, | |
| "eval_samples_per_second": 14.921, | |
| "eval_steps_per_second": 1.897, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.1430274135876043, | |
| "grad_norm": 0.5345009565353394, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 2.2879, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1573301549463647, | |
| "grad_norm": 1.6660727262496948, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 2.1803, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.17163289630512515, | |
| "grad_norm": 0.7011714577674866, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 2.3269, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.17163289630512515, | |
| "eval_loss": 2.3226821422576904, | |
| "eval_runtime": 23.7388, | |
| "eval_samples_per_second": 14.912, | |
| "eval_steps_per_second": 1.896, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.18593563766388557, | |
| "grad_norm": 0.861555814743042, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 2.3854, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.20023837902264602, | |
| "grad_norm": 0.7908144593238831, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 2.4378, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.21454112038140644, | |
| "grad_norm": 0.7242131233215332, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 2.4475, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.21454112038140644, | |
| "eval_loss": 2.2817511558532715, | |
| "eval_runtime": 23.748, | |
| "eval_samples_per_second": 14.906, | |
| "eval_steps_per_second": 1.895, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.22884386174016685, | |
| "grad_norm": 0.6113729476928711, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 2.2601, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.2431466030989273, | |
| "grad_norm": 0.6754246950149536, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 2.1681, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.25744934445768775, | |
| "grad_norm": 0.6604123115539551, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 2.2141, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.25744934445768775, | |
| "eval_loss": 2.253026247024536, | |
| "eval_runtime": 23.7486, | |
| "eval_samples_per_second": 14.906, | |
| "eval_steps_per_second": 1.895, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.27175208581644816, | |
| "grad_norm": 0.7334719300270081, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 2.2945, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.2860548271752086, | |
| "grad_norm": 0.6848142147064209, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 2.3294, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.300357568533969, | |
| "grad_norm": 0.6209741234779358, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 2.09, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.300357568533969, | |
| "eval_loss": 2.230947732925415, | |
| "eval_runtime": 23.7116, | |
| "eval_samples_per_second": 14.929, | |
| "eval_steps_per_second": 1.898, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.3146603098927294, | |
| "grad_norm": 0.699060320854187, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 2.189, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.3289630512514899, | |
| "grad_norm": 0.6675814390182495, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 2.2713, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.3432657926102503, | |
| "grad_norm": 0.6715078353881836, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 2.1175, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.3432657926102503, | |
| "eval_loss": 2.2183046340942383, | |
| "eval_runtime": 23.7149, | |
| "eval_samples_per_second": 14.927, | |
| "eval_steps_per_second": 1.898, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.3575685339690107, | |
| "grad_norm": 0.7805127501487732, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 2.1241, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.37187127532777114, | |
| "grad_norm": 0.7199813723564148, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 2.1669, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.38617401668653156, | |
| "grad_norm": 0.8319717049598694, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 2.1235, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.38617401668653156, | |
| "eval_loss": 2.2104547023773193, | |
| "eval_runtime": 23.7384, | |
| "eval_samples_per_second": 14.913, | |
| "eval_steps_per_second": 1.896, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.40047675804529204, | |
| "grad_norm": 0.7010225057601929, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 2.3046, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.41477949940405245, | |
| "grad_norm": 0.5844500064849854, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 2.1629, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.42908224076281287, | |
| "grad_norm": 0.7447096109390259, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 2.0827, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.42908224076281287, | |
| "eval_loss": 2.2069106101989746, | |
| "eval_runtime": 23.7385, | |
| "eval_samples_per_second": 14.912, | |
| "eval_steps_per_second": 1.896, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4433849821215733, | |
| "grad_norm": 0.8038536906242371, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 2.2257, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.4576877234803337, | |
| "grad_norm": 0.6432086229324341, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 2.1321, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.4719904648390942, | |
| "grad_norm": 0.6721349358558655, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 2.1555, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.4719904648390942, | |
| "eval_loss": 2.206080198287964, | |
| "eval_runtime": 23.7434, | |
| "eval_samples_per_second": 14.909, | |
| "eval_steps_per_second": 1.895, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.3843843396927488e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |