| { | |
| "best_metric": 0.5937426090240479, | |
| "best_model_checkpoint": "mikhail_panzo/zlm_b64_le5_s12000/checkpoint-500", | |
| "epoch": 0.418848167539267, | |
| "eval_steps": 500, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.041884816753926704, | |
| "grad_norm": 12.821951866149902, | |
| "learning_rate": 2.4000000000000003e-07, | |
| "loss": 1.1152, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.08376963350785341, | |
| "grad_norm": 6.186221122741699, | |
| "learning_rate": 4.900000000000001e-07, | |
| "loss": 1.0584, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1256544502617801, | |
| "grad_norm": 13.08674430847168, | |
| "learning_rate": 7.4e-07, | |
| "loss": 0.9996, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.16753926701570682, | |
| "grad_norm": 4.10632848739624, | |
| "learning_rate": 9.9e-07, | |
| "loss": 0.9355, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2094240837696335, | |
| "grad_norm": 3.2428064346313477, | |
| "learning_rate": 1.2400000000000002e-06, | |
| "loss": 0.852, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.2513089005235602, | |
| "grad_norm": 2.0861401557922363, | |
| "learning_rate": 1.4900000000000001e-06, | |
| "loss": 0.8241, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.2931937172774869, | |
| "grad_norm": 8.602317810058594, | |
| "learning_rate": 1.74e-06, | |
| "loss": 0.7629, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.33507853403141363, | |
| "grad_norm": 2.7594990730285645, | |
| "learning_rate": 1.9900000000000004e-06, | |
| "loss": 0.7418, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.3769633507853403, | |
| "grad_norm": 2.7047348022460938, | |
| "learning_rate": 2.24e-06, | |
| "loss": 0.7166, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.418848167539267, | |
| "grad_norm": 2.0666158199310303, | |
| "learning_rate": 2.4900000000000003e-06, | |
| "loss": 0.6876, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.418848167539267, | |
| "eval_loss": 0.5937426090240479, | |
| "eval_runtime": 263.0955, | |
| "eval_samples_per_second": 32.266, | |
| "eval_steps_per_second": 4.037, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 12000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 11, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4503134375055360.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |