| { | |
| "best_metric": 0.47028860449790955, | |
| "best_model_checkpoint": "data/Mistral-7B_task-2_120-samples_config-2_auto/checkpoint-27", | |
| "epoch": 12.0, | |
| "eval_steps": 500, | |
| "global_step": 66, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": 1.722359299659729, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.8072, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 1.4123291969299316, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.7321, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 1.450860619544983, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.7975, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "eval_loss": 0.7186066508293152, | |
| "eval_runtime": 30.8184, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 1.239404320716858, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.744, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 1.096447467803955, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.6684, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 0.8220383524894714, | |
| "learning_rate": 4e-05, | |
| "loss": 0.6495, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.5790643095970154, | |
| "eval_runtime": 30.8137, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 0.5528280735015869, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.5443, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": 0.6220946311950684, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.5733, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "grad_norm": 0.6053640246391296, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 0.5185, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "eval_loss": 0.5305019021034241, | |
| "eval_runtime": 30.8123, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 3.2727272727272725, | |
| "grad_norm": 0.53728848695755, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.5005, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 0.48586469888687134, | |
| "learning_rate": 8e-05, | |
| "loss": 0.4658, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.5516945719718933, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.4603, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.4857640266418457, | |
| "eval_runtime": 30.816, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.363636363636363, | |
| "grad_norm": 0.5344758629798889, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.4267, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 0.5485618710517883, | |
| "learning_rate": 9.999512620046522e-05, | |
| "loss": 0.4051, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.909090909090909, | |
| "eval_loss": 0.47028860449790955, | |
| "eval_runtime": 30.8101, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 5.090909090909091, | |
| "grad_norm": 0.5106642246246338, | |
| "learning_rate": 9.995614150494293e-05, | |
| "loss": 0.378, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.454545454545454, | |
| "grad_norm": 0.5522305369377136, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.3372, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.818181818181818, | |
| "grad_norm": 0.572932779788971, | |
| "learning_rate": 9.976136999909156e-05, | |
| "loss": 0.334, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.47919464111328125, | |
| "eval_runtime": 30.812, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 6.181818181818182, | |
| "grad_norm": 0.5302502512931824, | |
| "learning_rate": 9.96057350657239e-05, | |
| "loss": 0.2971, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.545454545454545, | |
| "grad_norm": 0.5685491561889648, | |
| "learning_rate": 9.941141907232765e-05, | |
| "loss": 0.2696, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "grad_norm": 0.6408573985099792, | |
| "learning_rate": 9.917857354066931e-05, | |
| "loss": 0.2396, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "eval_loss": 0.51518315076828, | |
| "eval_runtime": 30.8202, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 7.2727272727272725, | |
| "grad_norm": 0.6412529349327087, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.2035, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.636363636363637, | |
| "grad_norm": 0.7295396327972412, | |
| "learning_rate": 9.859805002892732e-05, | |
| "loss": 0.1572, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.8271765112876892, | |
| "learning_rate": 9.825082472361557e-05, | |
| "loss": 0.1658, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.5793466567993164, | |
| "eval_runtime": 30.8173, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.363636363636363, | |
| "grad_norm": 0.6950414180755615, | |
| "learning_rate": 9.786597487660337e-05, | |
| "loss": 0.1, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.727272727272727, | |
| "grad_norm": 0.6681713461875916, | |
| "learning_rate": 9.744380058222483e-05, | |
| "loss": 0.083, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.909090909090908, | |
| "eval_loss": 0.6896577477455139, | |
| "eval_runtime": 30.8157, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 9.090909090909092, | |
| "grad_norm": 1.0224828720092773, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.0937, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 9.454545454545455, | |
| "grad_norm": 0.679814875125885, | |
| "learning_rate": 9.648882429441257e-05, | |
| "loss": 0.0424, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.818181818181818, | |
| "grad_norm": 0.7080846428871155, | |
| "learning_rate": 9.595676696276172e-05, | |
| "loss": 0.05, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.7636191844940186, | |
| "eval_runtime": 30.8148, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 10.181818181818182, | |
| "grad_norm": 0.46838316321372986, | |
| "learning_rate": 9.538887392664544e-05, | |
| "loss": 0.0297, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 10.545454545454545, | |
| "grad_norm": 0.6797084808349609, | |
| "learning_rate": 9.478558801197065e-05, | |
| "loss": 0.0245, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "grad_norm": 0.8882114291191101, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.0314, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "eval_loss": 0.8598381876945496, | |
| "eval_runtime": 30.8142, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 11.272727272727273, | |
| "grad_norm": 0.7864226698875427, | |
| "learning_rate": 9.347474647526095e-05, | |
| "loss": 0.0208, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 11.636363636363637, | |
| "grad_norm": 0.7724862694740295, | |
| "learning_rate": 9.276821300802534e-05, | |
| "loss": 0.0194, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.7500962615013123, | |
| "learning_rate": 9.202833017478422e-05, | |
| "loss": 0.0226, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.804558277130127, | |
| "eval_runtime": 30.8177, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "step": 66, | |
| "total_flos": 1.2895031554945843e+17, | |
| "train_loss": 0.3158463573817051, | |
| "train_runtime": 4673.5222, | |
| "train_samples_per_second": 0.941, | |
| "train_steps_per_second": 0.053 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 250, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 7, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.2895031554945843e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |