| { | |
| "best_metric": 0.4513999819755554, | |
| "best_model_checkpoint": "mikhail_panzo/zlm_b128_le4_s12000/checkpoint-500", | |
| "epoch": 0.837696335078534, | |
| "eval_steps": 500, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08376963350785341, | |
| "grad_norm": 2.469545841217041, | |
| "learning_rate": 2.4500000000000003e-06, | |
| "loss": 1.0402, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.16753926701570682, | |
| "grad_norm": 3.3642780780792236, | |
| "learning_rate": 4.950000000000001e-06, | |
| "loss": 0.8472, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2513089005235602, | |
| "grad_norm": 1.8375087976455688, | |
| "learning_rate": 7.45e-06, | |
| "loss": 0.7331, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.33507853403141363, | |
| "grad_norm": 3.700824737548828, | |
| "learning_rate": 9.950000000000001e-06, | |
| "loss": 0.6472, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.418848167539267, | |
| "grad_norm": 1.7632888555526733, | |
| "learning_rate": 1.2450000000000001e-05, | |
| "loss": 0.6112, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.5026178010471204, | |
| "grad_norm": 3.1861908435821533, | |
| "learning_rate": 1.4950000000000001e-05, | |
| "loss": 0.5865, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5863874345549738, | |
| "grad_norm": 2.2845046520233154, | |
| "learning_rate": 1.745e-05, | |
| "loss": 0.5682, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.6701570680628273, | |
| "grad_norm": 2.079210042953491, | |
| "learning_rate": 1.995e-05, | |
| "loss": 0.5465, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7539267015706806, | |
| "grad_norm": 3.669891119003296, | |
| "learning_rate": 2.245e-05, | |
| "loss": 0.5302, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.837696335078534, | |
| "grad_norm": 2.4679417610168457, | |
| "learning_rate": 2.495e-05, | |
| "loss": 0.5137, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.837696335078534, | |
| "eval_loss": 0.4513999819755554, | |
| "eval_runtime": 277.1074, | |
| "eval_samples_per_second": 30.634, | |
| "eval_steps_per_second": 3.832, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 12000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 21, | |
| "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": 8985136057096320.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |