| { | |
| "best_metric": 1.0340019464492798, | |
| "best_model_checkpoint": "data/Llama-31-8B_task-2_120-samples_config-2_full/checkpoint-71", | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 110, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": 0.5546463131904602, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.6095, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 0.49554750323295593, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.5445, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.46318522095680237, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.6102, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "eval_loss": 1.574345588684082, | |
| "eval_runtime": 24.3664, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 0.46896636486053467, | |
| "learning_rate": 2.4e-05, | |
| "loss": 1.582, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 0.5056357979774475, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 1.5453, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 0.39398133754730225, | |
| "learning_rate": 4e-05, | |
| "loss": 1.5449, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.4670957326889038, | |
| "eval_runtime": 24.3579, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 0.3227110803127289, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.4329, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": 0.33419129252433777, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 1.422, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "grad_norm": 0.24367345869541168, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 1.3801, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "eval_loss": 1.3715490102767944, | |
| "eval_runtime": 24.3614, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 3.2727272727272725, | |
| "grad_norm": 0.25076785683631897, | |
| "learning_rate": 7.2e-05, | |
| "loss": 1.3492, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 0.25506436824798584, | |
| "learning_rate": 8e-05, | |
| "loss": 1.2881, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.22455543279647827, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 1.272, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.2586787939071655, | |
| "eval_runtime": 24.3581, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.363636363636363, | |
| "grad_norm": 0.2321537286043167, | |
| "learning_rate": 9.6e-05, | |
| "loss": 1.2144, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 0.2969757616519928, | |
| "learning_rate": 9.999512620046522e-05, | |
| "loss": 1.1581, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.909090909090909, | |
| "eval_loss": 1.1533732414245605, | |
| "eval_runtime": 24.3534, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 5.090909090909091, | |
| "grad_norm": 0.2301207035779953, | |
| "learning_rate": 9.995614150494293e-05, | |
| "loss": 1.1269, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.454545454545454, | |
| "grad_norm": 0.21130912005901337, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.0756, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.818181818181818, | |
| "grad_norm": 0.15461936593055725, | |
| "learning_rate": 9.976136999909156e-05, | |
| "loss": 1.0495, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.093610167503357, | |
| "eval_runtime": 24.3674, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 6.181818181818182, | |
| "grad_norm": 0.13114210963249207, | |
| "learning_rate": 9.96057350657239e-05, | |
| "loss": 1.045, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.545454545454545, | |
| "grad_norm": 0.12750868499279022, | |
| "learning_rate": 9.941141907232765e-05, | |
| "loss": 1.0173, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "grad_norm": 0.1173694059252739, | |
| "learning_rate": 9.917857354066931e-05, | |
| "loss": 1.0426, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "eval_loss": 1.076851725578308, | |
| "eval_runtime": 24.3562, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 7.2727272727272725, | |
| "grad_norm": 0.12811259925365448, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 1.0323, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.636363636363637, | |
| "grad_norm": 0.11960847675800323, | |
| "learning_rate": 9.859805002892732e-05, | |
| "loss": 0.9858, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.11962142586708069, | |
| "learning_rate": 9.825082472361557e-05, | |
| "loss": 1.0149, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.0612627267837524, | |
| "eval_runtime": 24.3609, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.363636363636363, | |
| "grad_norm": 0.13360188901424408, | |
| "learning_rate": 9.786597487660337e-05, | |
| "loss": 0.98, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.727272727272727, | |
| "grad_norm": 0.15958157181739807, | |
| "learning_rate": 9.744380058222483e-05, | |
| "loss": 0.9765, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.909090909090908, | |
| "eval_loss": 1.0515578985214233, | |
| "eval_runtime": 24.3612, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 9.090909090909092, | |
| "grad_norm": 0.1226094663143158, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.9969, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 9.454545454545455, | |
| "grad_norm": 0.13600100576877594, | |
| "learning_rate": 9.648882429441257e-05, | |
| "loss": 0.945, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.818181818181818, | |
| "grad_norm": 0.14928992092609406, | |
| "learning_rate": 9.595676696276172e-05, | |
| "loss": 0.9857, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.0432137250900269, | |
| "eval_runtime": 24.3577, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 10.181818181818182, | |
| "grad_norm": 0.12853661179542542, | |
| "learning_rate": 9.538887392664544e-05, | |
| "loss": 0.9367, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 10.545454545454545, | |
| "grad_norm": 0.2052777111530304, | |
| "learning_rate": 9.478558801197065e-05, | |
| "loss": 0.9345, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "grad_norm": 0.1295970231294632, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.9438, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "eval_loss": 1.0381237268447876, | |
| "eval_runtime": 24.3568, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 11.272727272727273, | |
| "grad_norm": 0.16599661111831665, | |
| "learning_rate": 9.347474647526095e-05, | |
| "loss": 0.9265, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 11.636363636363637, | |
| "grad_norm": 0.15103670954704285, | |
| "learning_rate": 9.276821300802534e-05, | |
| "loss": 0.9335, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.2108156383037567, | |
| "learning_rate": 9.202833017478422e-05, | |
| "loss": 0.9103, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.0346513986587524, | |
| "eval_runtime": 24.3619, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 12.363636363636363, | |
| "grad_norm": 0.160989910364151, | |
| "learning_rate": 9.125567491391476e-05, | |
| "loss": 0.9001, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 12.727272727272727, | |
| "grad_norm": 0.19350972771644592, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.8842, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 12.909090909090908, | |
| "eval_loss": 1.0340019464492798, | |
| "eval_runtime": 24.3565, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 13.090909090909092, | |
| "grad_norm": 0.2236265391111374, | |
| "learning_rate": 8.961448216775954e-05, | |
| "loss": 0.8737, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 13.454545454545455, | |
| "grad_norm": 0.17474591732025146, | |
| "learning_rate": 8.874722443520899e-05, | |
| "loss": 0.9117, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 13.818181818181818, | |
| "grad_norm": 0.19857965409755707, | |
| "learning_rate": 8.784975278258783e-05, | |
| "loss": 0.872, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 1.035847544670105, | |
| "eval_runtime": 24.3572, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 14.181818181818182, | |
| "grad_norm": 0.20292682945728302, | |
| "learning_rate": 8.692276703129421e-05, | |
| "loss": 0.8208, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 14.545454545454545, | |
| "grad_norm": 0.21912775933742523, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.8805, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 14.909090909090908, | |
| "grad_norm": 0.28970637917518616, | |
| "learning_rate": 8.498316702566828e-05, | |
| "loss": 0.7743, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 14.909090909090908, | |
| "eval_loss": 1.047881007194519, | |
| "eval_runtime": 24.3527, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 15.272727272727273, | |
| "grad_norm": 0.290974885225296, | |
| "learning_rate": 8.397206521307584e-05, | |
| "loss": 0.8288, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 15.636363636363637, | |
| "grad_norm": 0.3443286418914795, | |
| "learning_rate": 8.293447300593402e-05, | |
| "loss": 0.7872, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.35764703154563904, | |
| "learning_rate": 8.18711994874345e-05, | |
| "loss": 0.7851, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 1.0548561811447144, | |
| "eval_runtime": 24.3548, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 16.363636363636363, | |
| "grad_norm": 0.25758710503578186, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.8279, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 16.727272727272727, | |
| "grad_norm": 0.50641930103302, | |
| "learning_rate": 7.967094433018508e-05, | |
| "loss": 0.7288, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 16.90909090909091, | |
| "eval_loss": 1.0722235441207886, | |
| "eval_runtime": 24.3541, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 17.09090909090909, | |
| "grad_norm": 0.4707852602005005, | |
| "learning_rate": 7.85356783842216e-05, | |
| "loss": 0.7041, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 17.454545454545453, | |
| "grad_norm": 0.41037577390670776, | |
| "learning_rate": 7.737816117462752e-05, | |
| "loss": 0.7053, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 17.818181818181817, | |
| "grad_norm": 0.4197547137737274, | |
| "learning_rate": 7.619929529850397e-05, | |
| "loss": 0.6647, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 1.096216082572937, | |
| "eval_runtime": 24.3514, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 18.181818181818183, | |
| "grad_norm": 0.4885706305503845, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.7098, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 18.545454545454547, | |
| "grad_norm": 1.0484837293624878, | |
| "learning_rate": 7.378121045351378e-05, | |
| "loss": 0.6423, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 18.90909090909091, | |
| "grad_norm": 0.6320416927337646, | |
| "learning_rate": 7.254387703447154e-05, | |
| "loss": 0.6477, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 18.90909090909091, | |
| "eval_loss": 1.1323298215866089, | |
| "eval_runtime": 24.3555, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 19.272727272727273, | |
| "grad_norm": 0.8023173809051514, | |
| "learning_rate": 7.128896457825364e-05, | |
| "loss": 0.6205, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 19.636363636363637, | |
| "grad_norm": 0.564225971698761, | |
| "learning_rate": 7.001745162784477e-05, | |
| "loss": 0.5865, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.6270914077758789, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 0.6269, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 1.1613657474517822, | |
| "eval_runtime": 24.3619, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 110, | |
| "total_flos": 1.672905103549399e+17, | |
| "train_loss": 1.0003309444947677, | |
| "train_runtime": 6009.0646, | |
| "train_samples_per_second": 0.732, | |
| "train_steps_per_second": 0.042 | |
| } | |
| ], | |
| "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": 1.672905103549399e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |