| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 30, | |
| "global_step": 225, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.044444444444444446, | |
| "grad_norm": 0.3667745590209961, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.307, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08888888888888889, | |
| "grad_norm": 0.016515299677848816, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.0062, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.13333333333333333, | |
| "grad_norm": 0.09039086103439331, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.0042, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.13333333333333333, | |
| "eval_loss": 0.004083628766238689, | |
| "eval_runtime": 51.9939, | |
| "eval_samples_per_second": 9.617, | |
| "eval_steps_per_second": 0.481, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.17777777777777778, | |
| "grad_norm": 0.0036161942407488823, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.0041, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2222222222222222, | |
| "grad_norm": 0.03548239916563034, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.0039, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.26666666666666666, | |
| "grad_norm": 0.03349478542804718, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.004, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.26666666666666666, | |
| "eval_loss": 0.003905750811100006, | |
| "eval_runtime": 52.0248, | |
| "eval_samples_per_second": 9.611, | |
| "eval_steps_per_second": 0.481, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.3111111111111111, | |
| "grad_norm": 0.003334084525704384, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.004, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.35555555555555557, | |
| "grad_norm": 0.0032222650479525328, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.004, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.009061365388333797, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.0039, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.0039283959195017815, | |
| "eval_runtime": 52.0001, | |
| "eval_samples_per_second": 9.615, | |
| "eval_steps_per_second": 0.481, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 0.001912094303406775, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.004, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4888888888888889, | |
| "grad_norm": 0.007684116251766682, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.0039, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.5333333333333333, | |
| "grad_norm": 0.0023366182576864958, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 0.0039, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.5333333333333333, | |
| "eval_loss": 0.003954995423555374, | |
| "eval_runtime": 52.0391, | |
| "eval_samples_per_second": 9.608, | |
| "eval_steps_per_second": 0.48, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.5777777777777777, | |
| "grad_norm": 0.003150626551359892, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 0.0039, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.6222222222222222, | |
| "grad_norm": 0.001225905492901802, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.0039, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 0.004419529344886541, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.0039, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "eval_loss": 0.003903371747583151, | |
| "eval_runtime": 52.037, | |
| "eval_samples_per_second": 9.609, | |
| "eval_steps_per_second": 0.48, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "grad_norm": 0.0037097111344337463, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 0.0039, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.7555555555555555, | |
| "grad_norm": 0.011564449407160282, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.0039, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.019085539504885674, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0039, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.003919120877981186, | |
| "eval_runtime": 52.0957, | |
| "eval_samples_per_second": 9.598, | |
| "eval_steps_per_second": 0.48, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.8444444444444444, | |
| "grad_norm": 0.0039107827469706535, | |
| "learning_rate": 5.852620357053651e-06, | |
| "loss": 0.0039, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 0.011299680918455124, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.0039, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.9333333333333333, | |
| "grad_norm": 0.0018409871263429523, | |
| "learning_rate": 1.0926199633097157e-06, | |
| "loss": 0.0039, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.9333333333333333, | |
| "eval_loss": 0.003911687061190605, | |
| "eval_runtime": 52.1073, | |
| "eval_samples_per_second": 9.596, | |
| "eval_steps_per_second": 0.48, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.9777777777777777, | |
| "grad_norm": 0.0032750105019658804, | |
| "learning_rate": 1.2179748700879012e-07, | |
| "loss": 0.0038, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 225, | |
| "total_flos": 3.68076153035948e+17, | |
| "train_loss": 0.017504285780919924, | |
| "train_runtime": 1899.4331, | |
| "train_samples_per_second": 2.369, | |
| "train_steps_per_second": 0.118 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 225, | |
| "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": 3.68076153035948e+17, | |
| "train_batch_size": 10, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |