| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.4928909952606635, | |
| "eval_steps": 6, | |
| "global_step": 26, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.018957345971563982, | |
| "grad_norm": 1.21875, | |
| "learning_rate": 2e-05, | |
| "loss": 2.0442, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.018957345971563982, | |
| "eval_loss": 2.073391914367676, | |
| "eval_runtime": 6.8335, | |
| "eval_samples_per_second": 52.096, | |
| "eval_steps_per_second": 13.024, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.037914691943127965, | |
| "grad_norm": 1.40625, | |
| "learning_rate": 4e-05, | |
| "loss": 2.1817, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.05687203791469194, | |
| "grad_norm": 1.3125, | |
| "learning_rate": 6e-05, | |
| "loss": 2.0671, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.07582938388625593, | |
| "grad_norm": 1.5, | |
| "learning_rate": 8e-05, | |
| "loss": 1.9289, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0947867298578199, | |
| "grad_norm": 1.3984375, | |
| "learning_rate": 0.0001, | |
| "loss": 1.7705, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.11374407582938388, | |
| "grad_norm": 1.1171875, | |
| "learning_rate": 0.00012, | |
| "loss": 1.449, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.11374407582938388, | |
| "eval_loss": 1.2773768901824951, | |
| "eval_runtime": 7.1055, | |
| "eval_samples_per_second": 50.102, | |
| "eval_steps_per_second": 12.526, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.13270142180094788, | |
| "grad_norm": 1.125, | |
| "learning_rate": 0.00014, | |
| "loss": 1.3112, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.15165876777251186, | |
| "grad_norm": 1.4140625, | |
| "learning_rate": 0.00016, | |
| "loss": 1.2196, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.17061611374407584, | |
| "grad_norm": 0.87890625, | |
| "learning_rate": 0.00018, | |
| "loss": 1.1166, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.1895734597156398, | |
| "grad_norm": 0.8125, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0836, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.20853080568720378, | |
| "grad_norm": 0.4609375, | |
| "learning_rate": 0.00019997685019798912, | |
| "loss": 0.898, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.22748815165876776, | |
| "grad_norm": 0.376953125, | |
| "learning_rate": 0.00019990741151022301, | |
| "loss": 0.8548, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.22748815165876776, | |
| "eval_loss": 0.9005841612815857, | |
| "eval_runtime": 6.5385, | |
| "eval_samples_per_second": 54.447, | |
| "eval_steps_per_second": 13.612, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.24644549763033174, | |
| "grad_norm": 0.384765625, | |
| "learning_rate": 0.00019979171608653924, | |
| "loss": 0.8835, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.26540284360189575, | |
| "grad_norm": 0.326171875, | |
| "learning_rate": 0.00019962981749346078, | |
| "loss": 0.8374, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.2843601895734597, | |
| "grad_norm": 0.32421875, | |
| "learning_rate": 0.0001994217906893952, | |
| "loss": 0.8658, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.3033175355450237, | |
| "grad_norm": 0.326171875, | |
| "learning_rate": 0.000199167731989929, | |
| "loss": 0.8232, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.3222748815165877, | |
| "grad_norm": 0.234375, | |
| "learning_rate": 0.00019886775902323405, | |
| "loss": 0.7541, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.3412322274881517, | |
| "grad_norm": 0.2470703125, | |
| "learning_rate": 0.00019852201067560606, | |
| "loss": 0.8561, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.3412322274881517, | |
| "eval_loss": 0.7924264669418335, | |
| "eval_runtime": 7.7645, | |
| "eval_samples_per_second": 45.85, | |
| "eval_steps_per_second": 11.462, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.36018957345971564, | |
| "grad_norm": 0.267578125, | |
| "learning_rate": 0.00019813064702716094, | |
| "loss": 0.7909, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.3791469194312796, | |
| "grad_norm": 0.3046875, | |
| "learning_rate": 0.0001976938492777182, | |
| "loss": 0.7657, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3981042654028436, | |
| "grad_norm": 1.3984375, | |
| "learning_rate": 0.00019721181966290613, | |
| "loss": 0.7214, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.41706161137440756, | |
| "grad_norm": 0.33203125, | |
| "learning_rate": 0.00019668478136052774, | |
| "loss": 0.7511, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.43601895734597157, | |
| "grad_norm": 0.35546875, | |
| "learning_rate": 0.0001961129783872301, | |
| "loss": 0.7159, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.4549763033175355, | |
| "grad_norm": 0.279296875, | |
| "learning_rate": 0.00019549667548552556, | |
| "loss": 0.744, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.4549763033175355, | |
| "eval_loss": 0.717645525932312, | |
| "eval_runtime": 6.6339, | |
| "eval_samples_per_second": 53.664, | |
| "eval_steps_per_second": 13.416, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.47393364928909953, | |
| "grad_norm": 0.259765625, | |
| "learning_rate": 0.00019483615800121716, | |
| "loss": 0.6753, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.4928909952606635, | |
| "grad_norm": 0.26953125, | |
| "learning_rate": 0.00019413173175128473, | |
| "loss": 0.7068, | |
| "step": 26 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 156, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 26, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.817246469967053e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |