| { | |
| "best_metric": 1.9481738805770874, | |
| "best_model_checkpoint": "output/mashina-vremeni/checkpoint-69", | |
| "epoch": 3.0, | |
| "global_step": 69, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0001218137945423232, | |
| "loss": 2.4048, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 8.255708249541069e-05, | |
| "loss": 2.2664, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 3.703953841164296e-05, | |
| "loss": 2.056, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 5.679304716725937e-06, | |
| "loss": 2.0302, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.0518720149993896, | |
| "eval_runtime": 1.761, | |
| "eval_samples_per_second": 22.147, | |
| "eval_steps_per_second": 2.839, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 2.5438740879409643e-06, | |
| "loss": 1.8993, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 2.9039729902920112e-05, | |
| "loss": 1.8723, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 7.328142955681643e-05, | |
| "loss": 2.0132, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 0.00011542314562479964, | |
| "loss": 1.9871, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 0.00013656105589809228, | |
| "loss": 1.8905, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.9873172044754028, | |
| "eval_runtime": 1.7697, | |
| "eval_samples_per_second": 22.037, | |
| "eval_steps_per_second": 2.825, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 0.00012721317115188912, | |
| "loss": 1.8927, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 9.157274139492967e-05, | |
| "loss": 1.8758, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 4.5627258605070364e-05, | |
| "loss": 1.7925, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 9.986828848110892e-06, | |
| "loss": 1.8051, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.9481738805770874, | |
| "eval_runtime": 1.7668, | |
| "eval_samples_per_second": 22.074, | |
| "eval_steps_per_second": 2.83, | |
| "step": 69 | |
| } | |
| ], | |
| "max_steps": 69, | |
| "num_train_epochs": 3, | |
| "total_flos": 70940786688000.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |