| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9990610328638497, | |
| "eval_steps": 50, | |
| "global_step": 266, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03755868544600939, | |
| "grad_norm": 238.26105857074086, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 2.184, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07511737089201878, | |
| "grad_norm": 11.851124100947331, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.9583, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.11267605633802817, | |
| "grad_norm": 13.23709111000931, | |
| "learning_rate": 9.996112860009689e-06, | |
| "loss": 0.6666, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.15023474178403756, | |
| "grad_norm": 10.595263707766286, | |
| "learning_rate": 9.9271761563539e-06, | |
| "loss": 0.6073, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.18779342723004694, | |
| "grad_norm": 10.974358931394494, | |
| "learning_rate": 9.773228160797187e-06, | |
| "loss": 0.4888, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18779342723004694, | |
| "eval_loss": 0.40262532234191895, | |
| "eval_runtime": 84.8046, | |
| "eval_samples_per_second": 11.155, | |
| "eval_steps_per_second": 1.403, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.22535211267605634, | |
| "grad_norm": 6.999547671281271, | |
| "learning_rate": 9.536925023144742e-06, | |
| "loss": 0.332, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.26291079812206575, | |
| "grad_norm": 7.934709674256903, | |
| "learning_rate": 9.222343811959694e-06, | |
| "loss": 0.3059, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.3004694835680751, | |
| "grad_norm": 4.216247268345636, | |
| "learning_rate": 8.834912170647102e-06, | |
| "loss": 0.2967, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.3380281690140845, | |
| "grad_norm": 3.8491049580742125, | |
| "learning_rate": 8.38131467132416e-06, | |
| "loss": 0.3126, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3755868544600939, | |
| "grad_norm": 4.86769045072678, | |
| "learning_rate": 7.869377482205042e-06, | |
| "loss": 0.2158, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3755868544600939, | |
| "eval_loss": 0.2777794301509857, | |
| "eval_runtime": 84.5196, | |
| "eval_samples_per_second": 11.193, | |
| "eval_steps_per_second": 1.408, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4131455399061033, | |
| "grad_norm": 4.811792495045984, | |
| "learning_rate": 7.307933338397667e-06, | |
| "loss": 0.2894, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.4507042253521127, | |
| "grad_norm": 4.159739012234102, | |
| "learning_rate": 6.706669145845863e-06, | |
| "loss": 0.2692, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.48826291079812206, | |
| "grad_norm": 4.52034631499569, | |
| "learning_rate": 6.075958847790262e-06, | |
| "loss": 0.2533, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.5258215962441315, | |
| "grad_norm": 3.976938645835112, | |
| "learning_rate": 5.426684437395196e-06, | |
| "loss": 0.2145, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.5633802816901409, | |
| "grad_norm": 5.272788226812684, | |
| "learning_rate": 4.770048204709648e-06, | |
| "loss": 0.2311, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5633802816901409, | |
| "eval_loss": 0.21654193103313446, | |
| "eval_runtime": 84.4941, | |
| "eval_samples_per_second": 11.196, | |
| "eval_steps_per_second": 1.408, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.6009389671361502, | |
| "grad_norm": 4.509328497992535, | |
| "learning_rate": 4.1173794573691e-06, | |
| "loss": 0.2118, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.6384976525821596, | |
| "grad_norm": 3.367119100278668, | |
| "learning_rate": 3.479939049792817e-06, | |
| "loss": 0.2152, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.676056338028169, | |
| "grad_norm": 2.4731294441327822, | |
| "learning_rate": 2.8687250934422774e-06, | |
| "loss": 0.2025, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.7136150234741784, | |
| "grad_norm": 3.5116971601898213, | |
| "learning_rate": 2.2942832003289823e-06, | |
| "loss": 0.1917, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.7511737089201878, | |
| "grad_norm": 2.8679194958187146, | |
| "learning_rate": 1.7665245337452368e-06, | |
| "loss": 0.1806, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7511737089201878, | |
| "eval_loss": 0.18289822340011597, | |
| "eval_runtime": 84.4823, | |
| "eval_samples_per_second": 11.198, | |
| "eval_steps_per_second": 1.409, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7887323943661971, | |
| "grad_norm": 8.060687194002183, | |
| "learning_rate": 1.2945548054891322e-06, | |
| "loss": 0.1808, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.8262910798122066, | |
| "grad_norm": 2.8917862492734168, | |
| "learning_rate": 8.865171699890835e-07, | |
| "loss": 0.1439, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.863849765258216, | |
| "grad_norm": 2.7979916522519823, | |
| "learning_rate": 5.494517259623478e-07, | |
| "loss": 0.1977, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.9014084507042254, | |
| "grad_norm": 3.1531141963354, | |
| "learning_rate": 2.8917404970305096e-07, | |
| "loss": 0.1996, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.9389671361502347, | |
| "grad_norm": 4.235066375032124, | |
| "learning_rate": 1.1017485573197151e-07, | |
| "loss": 0.183, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.9389671361502347, | |
| "eval_loss": 0.1692400723695755, | |
| "eval_runtime": 84.3942, | |
| "eval_samples_per_second": 11.209, | |
| "eval_steps_per_second": 1.41, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.9765258215962441, | |
| "grad_norm": 3.024511496452823, | |
| "learning_rate": 1.554251601833201e-08, | |
| "loss": 0.154, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.9990610328638497, | |
| "step": 266, | |
| "total_flos": 47252837498880.0, | |
| "train_loss": 0.36927685298417745, | |
| "train_runtime": 7961.1782, | |
| "train_samples_per_second": 1.069, | |
| "train_steps_per_second": 0.033 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 266, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 47252837498880.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |