| { | |
| "best_metric": 1.5348279476165771, | |
| "best_model_checkpoint": "../saved_models_new/gptNEO_author_RB_epochs15_lr5e-05/checkpoint-4000", | |
| "epoch": 14.934660858743, | |
| "global_step": 24000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 1.6896, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_loss": 1.5621881484985352, | |
| "eval_runtime": 14.0653, | |
| "eval_samples_per_second": 25.382, | |
| "eval_steps_per_second": 12.726, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 4.7630893153281216e-05, | |
| "loss": 1.3351, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "eval_loss": 1.5348279476165771, | |
| "eval_runtime": 14.063, | |
| "eval_samples_per_second": 25.386, | |
| "eval_steps_per_second": 12.728, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 4.289267945984364e-05, | |
| "loss": 1.154, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "eval_loss": 1.5707106590270996, | |
| "eval_runtime": 14.0614, | |
| "eval_samples_per_second": 25.389, | |
| "eval_steps_per_second": 12.73, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "learning_rate": 3.815446576640607e-05, | |
| "loss": 0.9265, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "eval_loss": 1.6517491340637207, | |
| "eval_runtime": 14.0589, | |
| "eval_samples_per_second": 25.393, | |
| "eval_steps_per_second": 12.732, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 6.22, | |
| "learning_rate": 3.341625207296849e-05, | |
| "loss": 0.6998, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 6.22, | |
| "eval_loss": 1.8329756259918213, | |
| "eval_runtime": 14.0588, | |
| "eval_samples_per_second": 25.393, | |
| "eval_steps_per_second": 12.732, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 7.47, | |
| "learning_rate": 2.867803837953092e-05, | |
| "loss": 0.5298, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 7.47, | |
| "eval_loss": 1.9407192468643188, | |
| "eval_runtime": 14.0594, | |
| "eval_samples_per_second": 25.392, | |
| "eval_steps_per_second": 12.732, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "learning_rate": 2.3939824686093343e-05, | |
| "loss": 0.391, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "eval_loss": 2.0445475578308105, | |
| "eval_runtime": 14.0538, | |
| "eval_samples_per_second": 25.402, | |
| "eval_steps_per_second": 12.737, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 9.96, | |
| "learning_rate": 1.920161099265577e-05, | |
| "loss": 0.2698, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 9.96, | |
| "eval_loss": 2.14052414894104, | |
| "eval_runtime": 14.0566, | |
| "eval_samples_per_second": 25.397, | |
| "eval_steps_per_second": 12.734, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "learning_rate": 1.4463397299218195e-05, | |
| "loss": 0.1802, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "eval_loss": 2.2577126026153564, | |
| "eval_runtime": 14.0589, | |
| "eval_samples_per_second": 25.393, | |
| "eval_steps_per_second": 12.732, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 12.45, | |
| "learning_rate": 9.725183605780622e-06, | |
| "loss": 0.1244, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 12.45, | |
| "eval_loss": 2.3212833404541016, | |
| "eval_runtime": 14.0654, | |
| "eval_samples_per_second": 25.381, | |
| "eval_steps_per_second": 12.726, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 13.69, | |
| "learning_rate": 4.986969912343047e-06, | |
| "loss": 0.0874, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 13.69, | |
| "eval_loss": 2.3692986965179443, | |
| "eval_runtime": 14.0588, | |
| "eval_samples_per_second": 25.393, | |
| "eval_steps_per_second": 12.732, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 14.93, | |
| "learning_rate": 2.4875621890547267e-07, | |
| "loss": 0.0666, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 14.93, | |
| "eval_loss": 2.398601531982422, | |
| "eval_runtime": 14.0655, | |
| "eval_samples_per_second": 25.381, | |
| "eval_steps_per_second": 12.726, | |
| "step": 24000 | |
| } | |
| ], | |
| "max_steps": 24105, | |
| "num_train_epochs": 15, | |
| "total_flos": 6267141886574592.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |