| { | |
| "best_metric": 0.22931724786758423, | |
| "best_model_checkpoint": "checkpoints/instrucode/low-shot-task-specific-500-ex/social_i_qa/checkpoint-75", | |
| "epoch": 6.0, | |
| "eval_steps": 500, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 7.333333333333333e-05, | |
| "loss": 4.8517, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_loss": 1.9629485607147217, | |
| "eval_runtime": 4.1824, | |
| "eval_samples_per_second": 23.91, | |
| "eval_steps_per_second": 3.108, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 1.2888, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.41052401065826416, | |
| "eval_runtime": 4.1752, | |
| "eval_samples_per_second": 23.951, | |
| "eval_steps_per_second": 3.114, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 6.000000000000001e-05, | |
| "loss": 0.4255, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_loss": 0.32185935974121094, | |
| "eval_runtime": 4.1821, | |
| "eval_samples_per_second": 23.911, | |
| "eval_steps_per_second": 3.108, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 5.333333333333333e-05, | |
| "loss": 0.2955, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.666666666666667e-05, | |
| "loss": 0.2552, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.26777762174606323, | |
| "eval_runtime": 4.1799, | |
| "eval_samples_per_second": 23.924, | |
| "eval_steps_per_second": 3.11, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 4e-05, | |
| "loss": 0.2144, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "eval_loss": 0.24417449533939362, | |
| "eval_runtime": 4.1595, | |
| "eval_samples_per_second": 24.042, | |
| "eval_steps_per_second": 3.125, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.1887, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.22931724786758423, | |
| "eval_runtime": 4.1585, | |
| "eval_samples_per_second": 24.047, | |
| "eval_steps_per_second": 3.126, | |
| "step": 75 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 120, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 1.244223306989568e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |