| { | |
| "best_metric": 2.3010876178741455, | |
| "best_model_checkpoint": "./outputs/checkpoint-1400", | |
| "epoch": 1.0200364298724955, | |
| "eval_steps": 100, | |
| "global_step": 1400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0002, | |
| "loss": 2.7399, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 2.6418099403381348, | |
| "eval_runtime": 347.6157, | |
| "eval_samples_per_second": 18.049, | |
| "eval_steps_per_second": 2.258, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6052, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 2.5918312072753906, | |
| "eval_runtime": 333.731, | |
| "eval_samples_per_second": 18.8, | |
| "eval_steps_per_second": 2.352, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5622, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 2.551574468612671, | |
| "eval_runtime": 204.9306, | |
| "eval_samples_per_second": 30.615, | |
| "eval_steps_per_second": 3.831, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5366, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 2.517575263977051, | |
| "eval_runtime": 204.3925, | |
| "eval_samples_per_second": 30.696, | |
| "eval_steps_per_second": 3.841, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4946, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 2.4924821853637695, | |
| "eval_runtime": 204.4035, | |
| "eval_samples_per_second": 30.694, | |
| "eval_steps_per_second": 3.84, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4686, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 2.4666266441345215, | |
| "eval_runtime": 207.3453, | |
| "eval_samples_per_second": 30.259, | |
| "eval_steps_per_second": 3.786, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4503, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 2.4440107345581055, | |
| "eval_runtime": 205.5485, | |
| "eval_samples_per_second": 30.523, | |
| "eval_steps_per_second": 3.819, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4271, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 2.4231719970703125, | |
| "eval_runtime": 204.3763, | |
| "eval_samples_per_second": 30.698, | |
| "eval_steps_per_second": 3.841, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3911, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 2.401764154434204, | |
| "eval_runtime": 204.7042, | |
| "eval_samples_per_second": 30.649, | |
| "eval_steps_per_second": 3.835, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3881, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 2.3822004795074463, | |
| "eval_runtime": 204.6048, | |
| "eval_samples_per_second": 30.664, | |
| "eval_steps_per_second": 3.837, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3811, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 2.360255479812622, | |
| "eval_runtime": 283.1616, | |
| "eval_samples_per_second": 22.157, | |
| "eval_steps_per_second": 2.772, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3378, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 2.341019868850708, | |
| "eval_runtime": 204.7618, | |
| "eval_samples_per_second": 30.64, | |
| "eval_steps_per_second": 3.834, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3496, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 2.326402187347412, | |
| "eval_runtime": 365.7801, | |
| "eval_samples_per_second": 17.152, | |
| "eval_steps_per_second": 2.146, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 2.3010876178741455, | |
| "eval_runtime": 204.3039, | |
| "eval_samples_per_second": 30.709, | |
| "eval_steps_per_second": 3.842, | |
| "step": 1400 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4116, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 4.091805541361664e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |