| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 6000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.5833333333333334e-05, | |
| "loss": 0.9647, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.020086001604795456, | |
| "eval_runtime": 364.5688, | |
| "eval_samples_per_second": 1.369, | |
| "eval_steps_per_second": 0.343, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.0251, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.8655614256858826, | |
| "eval_runtime": 369.0051, | |
| "eval_samples_per_second": 1.352, | |
| "eval_steps_per_second": 0.339, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.0319, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 2.7063565254211426, | |
| "eval_runtime": 366.5111, | |
| "eval_samples_per_second": 1.361, | |
| "eval_steps_per_second": 0.341, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.0234, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 3.704106569290161, | |
| "eval_runtime": 367.9633, | |
| "eval_samples_per_second": 1.356, | |
| "eval_steps_per_second": 0.34, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 2.916666666666667e-05, | |
| "loss": 0.0192, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_loss": 4.969351768493652, | |
| "eval_runtime": 368.4104, | |
| "eval_samples_per_second": 1.354, | |
| "eval_steps_per_second": 0.339, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0178, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_loss": 4.055324554443359, | |
| "eval_runtime": 365.6477, | |
| "eval_samples_per_second": 1.365, | |
| "eval_steps_per_second": 0.342, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.016, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_loss": 3.822715997695923, | |
| "eval_runtime": 363.9829, | |
| "eval_samples_per_second": 1.371, | |
| "eval_steps_per_second": 0.343, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0151, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 4.511150360107422, | |
| "eval_runtime": 364.6494, | |
| "eval_samples_per_second": 1.368, | |
| "eval_steps_per_second": 0.343, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.0144, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "eval_loss": 5.14383602142334, | |
| "eval_runtime": 367.2726, | |
| "eval_samples_per_second": 1.359, | |
| "eval_steps_per_second": 0.34, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.0137, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_loss": 3.8697893619537354, | |
| "eval_runtime": 365.2291, | |
| "eval_samples_per_second": 1.366, | |
| "eval_steps_per_second": 0.342, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 0.0126, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "eval_loss": 2.9502100944519043, | |
| "eval_runtime": 362.1735, | |
| "eval_samples_per_second": 1.378, | |
| "eval_steps_per_second": 0.345, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0114, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.815199851989746, | |
| "eval_runtime": 375.0727, | |
| "eval_samples_per_second": 1.33, | |
| "eval_steps_per_second": 0.333, | |
| "step": 6000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 6000, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 5077881623347200.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |