| { | |
| "best_metric": 0.4858732521533966, | |
| "best_model_checkpoint": "checkpoints/instrucode/low-shot-task-specific-500-ex/svamp/checkpoint-62", | |
| "epoch": 4.96, | |
| "eval_steps": 500, | |
| "global_step": 62, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 7.400000000000001e-05, | |
| "loss": 4.1509, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_loss": 2.4244415760040283, | |
| "eval_runtime": 4.2583, | |
| "eval_samples_per_second": 23.484, | |
| "eval_steps_per_second": 3.053, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 6.733333333333333e-05, | |
| "loss": 2.0378, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.0243476629257202, | |
| "eval_runtime": 4.2676, | |
| "eval_samples_per_second": 23.432, | |
| "eval_steps_per_second": 3.046, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 6.0666666666666666e-05, | |
| "loss": 1.1215, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_loss": 0.7239224314689636, | |
| "eval_runtime": 4.2565, | |
| "eval_samples_per_second": 23.494, | |
| "eval_steps_per_second": 3.054, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 5.4000000000000005e-05, | |
| "loss": 0.7816, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.7333333333333336e-05, | |
| "loss": 0.5969, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.5359882116317749, | |
| "eval_runtime": 4.2654, | |
| "eval_samples_per_second": 23.444, | |
| "eval_steps_per_second": 3.048, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 4.066666666666667e-05, | |
| "loss": 0.5028, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "eval_loss": 0.4858732521533966, | |
| "eval_runtime": 4.2509, | |
| "eval_samples_per_second": 23.524, | |
| "eval_steps_per_second": 3.058, | |
| "step": 62 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 120, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 1.050122493001728e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |