| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "global_step": 1671, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12, | |
| "eval_loss": 0.49560546875, | |
| "eval_runtime": 8.4829, | |
| "eval_samples_per_second": 63.657, | |
| "eval_steps_per_second": 8.016, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_loss": 0.479248046875, | |
| "eval_runtime": 8.8248, | |
| "eval_samples_per_second": 61.191, | |
| "eval_steps_per_second": 7.706, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 5e-06, | |
| "loss": 1.6261, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 0.47021484375, | |
| "eval_runtime": 7.5494, | |
| "eval_samples_per_second": 71.529, | |
| "eval_steps_per_second": 9.007, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_loss": 0.465087890625, | |
| "eval_runtime": 8.7604, | |
| "eval_samples_per_second": 61.641, | |
| "eval_steps_per_second": 7.762, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4577, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_loss": 0.46044921875, | |
| "eval_runtime": 7.9849, | |
| "eval_samples_per_second": 67.627, | |
| "eval_steps_per_second": 8.516, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_loss": 0.455322265625, | |
| "eval_runtime": 8.5578, | |
| "eval_samples_per_second": 63.1, | |
| "eval_steps_per_second": 7.946, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_loss": 0.45263671875, | |
| "eval_runtime": 7.841, | |
| "eval_samples_per_second": 68.868, | |
| "eval_steps_per_second": 8.672, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4486, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_loss": 0.450439453125, | |
| "eval_runtime": 10.8958, | |
| "eval_samples_per_second": 49.56, | |
| "eval_steps_per_second": 6.241, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 1671, | |
| "total_flos": 1.1778474718632346e+17, | |
| "train_loss": 0.8031303182039198, | |
| "train_runtime": 6338.6403, | |
| "train_samples_per_second": 8.433, | |
| "train_steps_per_second": 0.264 | |
| } | |
| ], | |
| "max_steps": 1671, | |
| "num_train_epochs": 1, | |
| "total_flos": 1.1778474718632346e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |