| { | |
| "best_metric": 0.7025482654571533, | |
| "best_model_checkpoint": "../../models/lora/codellama-34b-multitask_cot_reflection/checkpoint-200", | |
| "epoch": 25.0, | |
| "eval_steps": 200, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 2.9999999999999997e-05, | |
| "loss": 1.9595, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 5.9999999999999995e-05, | |
| "loss": 1.9443, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 8.999999999999999e-05, | |
| "loss": 1.7826, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 0.00011999999999999999, | |
| "loss": 1.4385, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 0.00015, | |
| "loss": 1.0892, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 0.9362, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "learning_rate": 0.00020999999999999998, | |
| "loss": 0.9101, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 0.8864, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.62, | |
| "learning_rate": 0.00027, | |
| "loss": 0.8796, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8521, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "learning_rate": 0.0002921052631578947, | |
| "loss": 0.7999, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "learning_rate": 0.0002842105263157894, | |
| "loss": 0.7828, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 0.0002763157894736842, | |
| "loss": 0.7696, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "learning_rate": 0.0002684210526315789, | |
| "loss": 0.7555, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "learning_rate": 0.0002605263157894737, | |
| "loss": 0.7609, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.00025263157894736836, | |
| "loss": 0.7509, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 10.62, | |
| "learning_rate": 0.00024473684210526314, | |
| "loss": 0.7587, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 11.25, | |
| "learning_rate": 0.00023684210526315788, | |
| "loss": 0.7453, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 11.88, | |
| "learning_rate": 0.00022894736842105263, | |
| "loss": 0.7222, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "learning_rate": 0.00022105263157894733, | |
| "loss": 0.7216, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "eval_loss": 0.7025482654571533, | |
| "eval_runtime": 3.0725, | |
| "eval_samples_per_second": 6.835, | |
| "eval_steps_per_second": 0.976, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 13.12, | |
| "learning_rate": 0.00021315789473684208, | |
| "loss": 0.7195, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 13.75, | |
| "learning_rate": 0.00020526315789473683, | |
| "loss": 0.6919, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 14.38, | |
| "learning_rate": 0.00019736842105263157, | |
| "loss": 0.7102, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 0.0001894736842105263, | |
| "loss": 0.6919, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 15.62, | |
| "learning_rate": 0.00018157894736842105, | |
| "loss": 0.7025, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 16.25, | |
| "learning_rate": 0.0001736842105263158, | |
| "loss": 0.6906, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 16.88, | |
| "learning_rate": 0.00016578947368421052, | |
| "loss": 0.667, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 17.5, | |
| "learning_rate": 0.00015789473684210524, | |
| "loss": 0.6554, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 18.12, | |
| "learning_rate": 0.00015, | |
| "loss": 0.6702, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 18.75, | |
| "learning_rate": 0.0001421052631578947, | |
| "loss": 0.6382, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 19.38, | |
| "learning_rate": 0.00013421052631578946, | |
| "loss": 0.6389, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.00012631578947368418, | |
| "loss": 0.6431, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 20.62, | |
| "learning_rate": 0.00011842105263157894, | |
| "loss": 0.6427, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 21.25, | |
| "learning_rate": 0.00011052631578947366, | |
| "loss": 0.6398, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 21.88, | |
| "learning_rate": 0.00010263157894736841, | |
| "loss": 0.609, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 22.5, | |
| "learning_rate": 9.473684210526315e-05, | |
| "loss": 0.6062, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 23.12, | |
| "learning_rate": 8.68421052631579e-05, | |
| "loss": 0.6085, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 23.75, | |
| "learning_rate": 7.894736842105262e-05, | |
| "loss": 0.5872, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 24.38, | |
| "learning_rate": 7.105263157894735e-05, | |
| "loss": 0.5877, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 6.315789473684209e-05, | |
| "loss": 0.5973, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 0.7764873504638672, | |
| "eval_runtime": 3.109, | |
| "eval_samples_per_second": 6.755, | |
| "eval_steps_per_second": 0.965, | |
| "step": 400 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 480, | |
| "num_train_epochs": 30, | |
| "save_steps": 200, | |
| "total_flos": 2.6530697187989914e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |