| { | |
| "best_metric": 2.852426052093506, | |
| "best_model_checkpoint": "./output/checkpoint-102", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 255, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.5936680271799274e-05, | |
| "loss": 3.0929, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 3.187336054359855e-05, | |
| "loss": 3.1544, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 4.781004081539782e-05, | |
| "loss": 3.002, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 6.37467210871971e-05, | |
| "loss": 2.952, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 7.968340135899636e-05, | |
| "loss": 2.8375, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.8777899742126465, | |
| "eval_runtime": 18.561, | |
| "eval_samples_per_second": 10.937, | |
| "eval_steps_per_second": 0.7, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 9.562008163079564e-05, | |
| "loss": 2.7301, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 0.00011155676190259492, | |
| "loss": 2.6083, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 0.0001274934421743942, | |
| "loss": 2.5973, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 0.00014343012244619346, | |
| "loss": 2.7028, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 0.00015936680271799272, | |
| "loss": 2.6389, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.852426052093506, | |
| "eval_runtime": 18.9267, | |
| "eval_samples_per_second": 10.726, | |
| "eval_steps_per_second": 0.687, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 0.000175303482989792, | |
| "loss": 2.3737, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 0.00019124016326159128, | |
| "loss": 2.4245, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 0.00020717684353339054, | |
| "loss": 2.522, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 0.00022311352380518983, | |
| "loss": 2.4049, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 0.00023905020407698907, | |
| "loss": 2.4488, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.864370346069336, | |
| "eval_runtime": 19.1132, | |
| "eval_samples_per_second": 10.621, | |
| "eval_steps_per_second": 0.68, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 0.0002549868843487884, | |
| "loss": 2.3544, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 0.0002709235646205876, | |
| "loss": 2.1872, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 0.0002868602448923869, | |
| "loss": 2.221, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 0.0003027969251641862, | |
| "loss": 2.4036, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 0.00031873360543598544, | |
| "loss": 2.4628, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 3.105098247528076, | |
| "eval_runtime": 19.2038, | |
| "eval_samples_per_second": 10.571, | |
| "eval_steps_per_second": 0.677, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 0.0003346702857077847, | |
| "loss": 2.2478, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 0.000350606965979584, | |
| "loss": 2.261, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "learning_rate": 0.00036654364625138324, | |
| "loss": 2.3157, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "learning_rate": 0.00038248032652318255, | |
| "loss": 2.1135, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 0.0003984170067949818, | |
| "loss": 2.2104, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 3.1024045944213867, | |
| "eval_runtime": 19.1911, | |
| "eval_samples_per_second": 10.578, | |
| "eval_steps_per_second": 0.677, | |
| "step": 255 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 255, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 2188457088743520.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |