| { | |
| "best_metric": 0.49923568964004517, | |
| "best_model_checkpoint": "mikhail-panzo/zlm_b64_le5_s8000/checkpoint-1000", | |
| "epoch": 0.837696335078534, | |
| "eval_steps": 500, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.041884816753926704, | |
| "grad_norm": 11.218673706054688, | |
| "learning_rate": 2.4500000000000004e-07, | |
| "loss": 1.1196, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.08376963350785341, | |
| "grad_norm": 5.423076629638672, | |
| "learning_rate": 4.95e-07, | |
| "loss": 1.057, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1256544502617801, | |
| "grad_norm": 12.713382720947266, | |
| "learning_rate": 7.4e-07, | |
| "loss": 1.0037, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.16753926701570682, | |
| "grad_norm": 4.434875965118408, | |
| "learning_rate": 9.9e-07, | |
| "loss": 0.9381, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2094240837696335, | |
| "grad_norm": 3.668553590774536, | |
| "learning_rate": 1.2400000000000002e-06, | |
| "loss": 0.8591, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.2513089005235602, | |
| "grad_norm": 2.2456483840942383, | |
| "learning_rate": 1.4900000000000001e-06, | |
| "loss": 0.8262, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.2931937172774869, | |
| "grad_norm": 2.70170259475708, | |
| "learning_rate": 1.74e-06, | |
| "loss": 0.7634, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.33507853403141363, | |
| "grad_norm": 2.9946064949035645, | |
| "learning_rate": 1.9900000000000004e-06, | |
| "loss": 0.7336, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.3769633507853403, | |
| "grad_norm": 2.2950398921966553, | |
| "learning_rate": 2.24e-06, | |
| "loss": 0.7121, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.418848167539267, | |
| "grad_norm": 8.007049560546875, | |
| "learning_rate": 2.4900000000000003e-06, | |
| "loss": 0.6879, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.418848167539267, | |
| "eval_loss": 0.5903987884521484, | |
| "eval_runtime": 258.509, | |
| "eval_samples_per_second": 32.838, | |
| "eval_steps_per_second": 4.108, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4607329842931937, | |
| "grad_norm": 3.0767672061920166, | |
| "learning_rate": 2.7400000000000004e-06, | |
| "loss": 0.6734, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.5026178010471204, | |
| "grad_norm": 8.150467872619629, | |
| "learning_rate": 2.99e-06, | |
| "loss": 0.6539, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.5445026178010471, | |
| "grad_norm": 2.2473220825195312, | |
| "learning_rate": 3.2400000000000003e-06, | |
| "loss": 0.657, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.5863874345549738, | |
| "grad_norm": 2.448143482208252, | |
| "learning_rate": 3.49e-06, | |
| "loss": 0.6258, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.6282722513089005, | |
| "grad_norm": 4.810636520385742, | |
| "learning_rate": 3.74e-06, | |
| "loss": 0.6155, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.6701570680628273, | |
| "grad_norm": 2.5927581787109375, | |
| "learning_rate": 3.990000000000001e-06, | |
| "loss": 0.6073, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.7120418848167539, | |
| "grad_norm": 1.913765549659729, | |
| "learning_rate": 4.24e-06, | |
| "loss": 0.6118, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.7539267015706806, | |
| "grad_norm": 2.210376262664795, | |
| "learning_rate": 4.49e-06, | |
| "loss": 0.5807, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.7958115183246073, | |
| "grad_norm": 3.323096752166748, | |
| "learning_rate": 4.74e-06, | |
| "loss": 0.5915, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.837696335078534, | |
| "grad_norm": 2.1455421447753906, | |
| "learning_rate": 4.9900000000000005e-06, | |
| "loss": 0.5606, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.837696335078534, | |
| "eval_loss": 0.49923568964004517, | |
| "eval_runtime": 260.1008, | |
| "eval_samples_per_second": 32.637, | |
| "eval_steps_per_second": 4.083, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 8000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "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": 8963407491426432.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |