| { | |
| "best_metric": 0.7995870113372803, | |
| "best_model_checkpoint": "data/Mistral-7B_task-1_120-samples_config-1_auto/checkpoint-33", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 110, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09090909090909091, | |
| "grad_norm": 3.058340549468994, | |
| "learning_rate": 1.818181818181818e-06, | |
| "loss": 1.0798, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": 2.7905619144439697, | |
| "learning_rate": 3.636363636363636e-06, | |
| "loss": 1.0715, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 2.621762752532959, | |
| "learning_rate": 7.272727272727272e-06, | |
| "loss": 1.0801, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.5454545454545454, | |
| "grad_norm": 3.1078500747680664, | |
| "learning_rate": 1.0909090909090909e-05, | |
| "loss": 1.1039, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 2.491581439971924, | |
| "learning_rate": 1.4545454545454545e-05, | |
| "loss": 0.9904, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": 2.5813634395599365, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 1.0952, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.9499928951263428, | |
| "eval_runtime": 9.9905, | |
| "eval_samples_per_second": 2.402, | |
| "eval_steps_per_second": 2.402, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 2.3418071269989014, | |
| "learning_rate": 2.1818181818181818e-05, | |
| "loss": 0.9362, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.2727272727272727, | |
| "grad_norm": 1.7050814628601074, | |
| "learning_rate": 2.5454545454545454e-05, | |
| "loss": 0.8387, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 1.4122742414474487, | |
| "learning_rate": 2.909090909090909e-05, | |
| "loss": 0.7974, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6363636363636362, | |
| "grad_norm": 1.5601708889007568, | |
| "learning_rate": 3.272727272727273e-05, | |
| "loss": 0.8938, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 1.7986031770706177, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 0.8813, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.438372254371643, | |
| "learning_rate": 4e-05, | |
| "loss": 0.738, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.8333826065063477, | |
| "eval_runtime": 9.9894, | |
| "eval_samples_per_second": 2.403, | |
| "eval_steps_per_second": 2.403, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 1.469726324081421, | |
| "learning_rate": 4.3636363636363636e-05, | |
| "loss": 0.6784, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.3636363636363638, | |
| "grad_norm": 1.2338181734085083, | |
| "learning_rate": 4.7272727272727275e-05, | |
| "loss": 0.7296, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": 1.4453539848327637, | |
| "learning_rate": 5.090909090909091e-05, | |
| "loss": 0.7257, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 1.4550710916519165, | |
| "learning_rate": 5.4545454545454546e-05, | |
| "loss": 0.693, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "grad_norm": 1.740180492401123, | |
| "learning_rate": 5.818181818181818e-05, | |
| "loss": 0.6322, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.7995870113372803, | |
| "eval_runtime": 9.9824, | |
| "eval_samples_per_second": 2.404, | |
| "eval_steps_per_second": 2.404, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.090909090909091, | |
| "grad_norm": 1.4883478879928589, | |
| "learning_rate": 6.181818181818182e-05, | |
| "loss": 0.6138, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.2727272727272725, | |
| "grad_norm": 1.2917379140853882, | |
| "learning_rate": 6.545454545454546e-05, | |
| "loss": 0.4963, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.4545454545454546, | |
| "grad_norm": 1.4996962547302246, | |
| "learning_rate": 6.90909090909091e-05, | |
| "loss": 0.5165, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 1.8155909776687622, | |
| "learning_rate": 7.272727272727273e-05, | |
| "loss": 0.5019, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.8181818181818183, | |
| "grad_norm": 1.5982214212417603, | |
| "learning_rate": 7.636363636363637e-05, | |
| "loss": 0.5005, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.7877131700515747, | |
| "learning_rate": 8e-05, | |
| "loss": 0.4414, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.8437328338623047, | |
| "eval_runtime": 9.9888, | |
| "eval_samples_per_second": 2.403, | |
| "eval_steps_per_second": 2.403, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.181818181818182, | |
| "grad_norm": 1.6378284692764282, | |
| "learning_rate": 8.363636363636364e-05, | |
| "loss": 0.3577, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.363636363636363, | |
| "grad_norm": 1.7813000679016113, | |
| "learning_rate": 8.727272727272727e-05, | |
| "loss": 0.3049, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.545454545454545, | |
| "grad_norm": 2.689056634902954, | |
| "learning_rate": 9.090909090909092e-05, | |
| "loss": 0.3101, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 2.8523430824279785, | |
| "learning_rate": 9.454545454545455e-05, | |
| "loss": 0.306, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 4.909090909090909, | |
| "grad_norm": 2.4803292751312256, | |
| "learning_rate": 9.818181818181818e-05, | |
| "loss": 0.2493, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.9666339755058289, | |
| "eval_runtime": 9.9909, | |
| "eval_samples_per_second": 2.402, | |
| "eval_steps_per_second": 2.402, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.090909090909091, | |
| "grad_norm": 1.6390739679336548, | |
| "learning_rate": 9.999899300364532e-05, | |
| "loss": 0.2011, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.2727272727272725, | |
| "grad_norm": 1.583099126815796, | |
| "learning_rate": 9.99909372761763e-05, | |
| "loss": 0.1846, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.454545454545454, | |
| "grad_norm": 2.01283597946167, | |
| "learning_rate": 9.997482711915927e-05, | |
| "loss": 0.1539, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.636363636363637, | |
| "grad_norm": 2.042282819747925, | |
| "learning_rate": 9.99506651282272e-05, | |
| "loss": 0.1331, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 5.818181818181818, | |
| "grad_norm": 2.783254861831665, | |
| "learning_rate": 9.991845519630678e-05, | |
| "loss": 0.1533, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.216506242752075, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.1348, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.2306486368179321, | |
| "eval_runtime": 9.9807, | |
| "eval_samples_per_second": 2.405, | |
| "eval_steps_per_second": 2.405, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.181818181818182, | |
| "grad_norm": 1.2987682819366455, | |
| "learning_rate": 9.982991356370404e-05, | |
| "loss": 0.0902, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.363636363636363, | |
| "grad_norm": 1.2827564477920532, | |
| "learning_rate": 9.977359612865423e-05, | |
| "loss": 0.0804, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.545454545454545, | |
| "grad_norm": 2.1345791816711426, | |
| "learning_rate": 9.970925928158274e-05, | |
| "loss": 0.0908, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 6.7272727272727275, | |
| "grad_norm": 2.1631665229797363, | |
| "learning_rate": 9.963691338830044e-05, | |
| "loss": 0.1026, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "grad_norm": 2.2595865726470947, | |
| "learning_rate": 9.955657010501806e-05, | |
| "loss": 0.1156, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.2281731367111206, | |
| "eval_runtime": 9.9804, | |
| "eval_samples_per_second": 2.405, | |
| "eval_steps_per_second": 2.405, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 7.090909090909091, | |
| "grad_norm": 1.5286589860916138, | |
| "learning_rate": 9.946824237646824e-05, | |
| "loss": 0.0793, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 7.2727272727272725, | |
| "grad_norm": 1.1395256519317627, | |
| "learning_rate": 9.937194443381972e-05, | |
| "loss": 0.0721, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 7.454545454545454, | |
| "grad_norm": 2.2062153816223145, | |
| "learning_rate": 9.926769179238466e-05, | |
| "loss": 0.071, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 7.636363636363637, | |
| "grad_norm": 2.096759796142578, | |
| "learning_rate": 9.915550124911866e-05, | |
| "loss": 0.0692, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 7.818181818181818, | |
| "grad_norm": 1.246720552444458, | |
| "learning_rate": 9.903539087991462e-05, | |
| "loss": 0.0788, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.8606150150299072, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.0976, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.3054982423782349, | |
| "eval_runtime": 9.9906, | |
| "eval_samples_per_second": 2.402, | |
| "eval_steps_per_second": 2.402, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 8.181818181818182, | |
| "grad_norm": 1.2230342626571655, | |
| "learning_rate": 9.877148934427037e-05, | |
| "loss": 0.0466, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 8.363636363636363, | |
| "grad_norm": 1.7322582006454468, | |
| "learning_rate": 9.862774069706346e-05, | |
| "loss": 0.0412, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 8.545454545454545, | |
| "grad_norm": 1.27932870388031, | |
| "learning_rate": 9.847615725553456e-05, | |
| "loss": 0.0605, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 8.727272727272727, | |
| "grad_norm": 4.297811508178711, | |
| "learning_rate": 9.831676344247342e-05, | |
| "loss": 0.0725, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 8.909090909090908, | |
| "grad_norm": 1.2094900608062744, | |
| "learning_rate": 9.814958493905963e-05, | |
| "loss": 0.0519, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 1.2516096830368042, | |
| "eval_runtime": 10.0, | |
| "eval_samples_per_second": 2.4, | |
| "eval_steps_per_second": 2.4, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 9.090909090909092, | |
| "grad_norm": 2.241234540939331, | |
| "learning_rate": 9.797464868072488e-05, | |
| "loss": 0.0537, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 9.272727272727273, | |
| "grad_norm": 0.9464065432548523, | |
| "learning_rate": 9.779198285281325e-05, | |
| "loss": 0.0269, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 9.454545454545455, | |
| "grad_norm": 2.2405903339385986, | |
| "learning_rate": 9.760161688604008e-05, | |
| "loss": 0.0461, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 9.636363636363637, | |
| "grad_norm": 1.0944693088531494, | |
| "learning_rate": 9.740358145174998e-05, | |
| "loss": 0.0452, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 9.818181818181818, | |
| "grad_norm": 1.5353902578353882, | |
| "learning_rate": 9.719790845697533e-05, | |
| "loss": 0.0367, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.6133581399917603, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.0399, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.3289762735366821, | |
| "eval_runtime": 9.9983, | |
| "eval_samples_per_second": 2.4, | |
| "eval_steps_per_second": 2.4, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 110, | |
| "total_flos": 1.4501983775883264e+16, | |
| "train_loss": 0.3803161793134429, | |
| "train_runtime": 1074.7679, | |
| "train_samples_per_second": 4.094, | |
| "train_steps_per_second": 0.512 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 550, | |
| "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": 1.4501983775883264e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |