| { |
| "best_global_step": 260, |
| "best_metric": 0.0018482182640582323, |
| "best_model_checkpoint": "./multitask_model/checkpoint-260", |
| "epoch": 4.0, |
| "eval_steps": 500, |
| "global_step": 260, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.15384615384615385, |
| "grad_norm": 8.8089017868042, |
| "learning_rate": 4.861538461538462e-05, |
| "loss": 2.1939, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.3076923076923077, |
| "grad_norm": 3.6326286792755127, |
| "learning_rate": 4.707692307692308e-05, |
| "loss": 1.1303, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.46153846153846156, |
| "grad_norm": 2.114581823348999, |
| "learning_rate": 4.553846153846154e-05, |
| "loss": 0.3031, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.6153846153846154, |
| "grad_norm": 0.4926183223724365, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.0667, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.7692307692307693, |
| "grad_norm": 0.26311376690864563, |
| "learning_rate": 4.2461538461538465e-05, |
| "loss": 0.0229, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.9230769230769231, |
| "grad_norm": 0.16266165673732758, |
| "learning_rate": 4.0923076923076925e-05, |
| "loss": 0.0119, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.005465179681777954, |
| "eval_runtime": 28.3814, |
| "eval_samples_per_second": 4.545, |
| "eval_steps_per_second": 0.599, |
| "step": 65 |
| }, |
| { |
| "epoch": 1.0769230769230769, |
| "grad_norm": 0.09443770349025726, |
| "learning_rate": 3.9384615384615384e-05, |
| "loss": 0.0088, |
| "step": 70 |
| }, |
| { |
| "epoch": 1.2307692307692308, |
| "grad_norm": 0.05983053520321846, |
| "learning_rate": 3.784615384615385e-05, |
| "loss": 0.0066, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.3846153846153846, |
| "grad_norm": 0.045136693865060806, |
| "learning_rate": 3.630769230769231e-05, |
| "loss": 0.0053, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.5384615384615383, |
| "grad_norm": 0.05789942666888237, |
| "learning_rate": 3.476923076923077e-05, |
| "loss": 0.0994, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.6923076923076923, |
| "grad_norm": 0.06267621368169785, |
| "learning_rate": 3.323076923076923e-05, |
| "loss": 0.0058, |
| "step": 110 |
| }, |
| { |
| "epoch": 1.8461538461538463, |
| "grad_norm": 0.050915952771902084, |
| "learning_rate": 3.1692307692307696e-05, |
| "loss": 0.0047, |
| "step": 120 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 0.0482671745121479, |
| "learning_rate": 3.0153846153846155e-05, |
| "loss": 0.0039, |
| "step": 130 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.0026640458963811398, |
| "eval_runtime": 28.7779, |
| "eval_samples_per_second": 4.483, |
| "eval_steps_per_second": 0.591, |
| "step": 130 |
| }, |
| { |
| "epoch": 2.1538461538461537, |
| "grad_norm": 0.041481729596853256, |
| "learning_rate": 2.8615384615384615e-05, |
| "loss": 0.0038, |
| "step": 140 |
| }, |
| { |
| "epoch": 2.3076923076923075, |
| "grad_norm": 0.04328610375523567, |
| "learning_rate": 2.7076923076923078e-05, |
| "loss": 0.0037, |
| "step": 150 |
| }, |
| { |
| "epoch": 2.4615384615384617, |
| "grad_norm": 0.03569851815700531, |
| "learning_rate": 2.5538461538461538e-05, |
| "loss": 0.0033, |
| "step": 160 |
| }, |
| { |
| "epoch": 2.6153846153846154, |
| "grad_norm": 0.027992915362119675, |
| "learning_rate": 2.4e-05, |
| "loss": 0.0033, |
| "step": 170 |
| }, |
| { |
| "epoch": 2.769230769230769, |
| "grad_norm": 0.026577744632959366, |
| "learning_rate": 2.246153846153846e-05, |
| "loss": 0.003, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.9230769230769234, |
| "grad_norm": 0.031246010214090347, |
| "learning_rate": 2.0923076923076923e-05, |
| "loss": 0.0029, |
| "step": 190 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.001924663782119751, |
| "eval_runtime": 29.0281, |
| "eval_samples_per_second": 4.444, |
| "eval_steps_per_second": 0.586, |
| "step": 195 |
| }, |
| { |
| "epoch": 3.076923076923077, |
| "grad_norm": 0.03536583110690117, |
| "learning_rate": 1.9384615384615383e-05, |
| "loss": 0.3434, |
| "step": 200 |
| }, |
| { |
| "epoch": 3.230769230769231, |
| "grad_norm": 0.03196291625499725, |
| "learning_rate": 1.7846153846153846e-05, |
| "loss": 0.0032, |
| "step": 210 |
| }, |
| { |
| "epoch": 3.3846153846153846, |
| "grad_norm": 0.05442598834633827, |
| "learning_rate": 1.630769230769231e-05, |
| "loss": 0.0034, |
| "step": 220 |
| }, |
| { |
| "epoch": 3.5384615384615383, |
| "grad_norm": 0.02654326893389225, |
| "learning_rate": 1.4769230769230772e-05, |
| "loss": 0.0766, |
| "step": 230 |
| }, |
| { |
| "epoch": 3.6923076923076925, |
| "grad_norm": 0.034997936338186264, |
| "learning_rate": 1.3230769230769233e-05, |
| "loss": 0.0038, |
| "step": 240 |
| }, |
| { |
| "epoch": 3.8461538461538463, |
| "grad_norm": 0.027517210692167282, |
| "learning_rate": 1.1692307692307693e-05, |
| "loss": 0.0041, |
| "step": 250 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 0.04248378798365593, |
| "learning_rate": 1.0153846153846154e-05, |
| "loss": 0.0027, |
| "step": 260 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.0018482182640582323, |
| "eval_runtime": 29.088, |
| "eval_samples_per_second": 4.435, |
| "eval_steps_per_second": 0.584, |
| "step": 260 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 325, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 0.0, |
| "train_batch_size": 8, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
