| { | |
| "best_metric": 0.46927714347839355, | |
| "best_model_checkpoint": "data/Mistral-7B_task-2_120-samples_config-1_auto/checkpoint-44", | |
| "epoch": 11.0, | |
| "eval_steps": 500, | |
| "global_step": 121, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09090909090909091, | |
| "grad_norm": 2.1914315223693848, | |
| "learning_rate": 1.818181818181818e-06, | |
| "loss": 0.779, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": 1.8220888376235962, | |
| "learning_rate": 3.636363636363636e-06, | |
| "loss": 0.8353, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 1.455124020576477, | |
| "learning_rate": 7.272727272727272e-06, | |
| "loss": 0.7281, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.5454545454545454, | |
| "grad_norm": 1.7466682195663452, | |
| "learning_rate": 1.0909090909090909e-05, | |
| "loss": 0.7833, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 1.527980923652649, | |
| "learning_rate": 1.4545454545454545e-05, | |
| "loss": 0.7654, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": 1.1736787557601929, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 0.6963, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.6489822268486023, | |
| "eval_runtime": 30.8118, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 1.0861847400665283, | |
| "learning_rate": 2.1818181818181818e-05, | |
| "loss": 0.6836, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.2727272727272727, | |
| "grad_norm": 0.7637216448783875, | |
| "learning_rate": 2.5454545454545454e-05, | |
| "loss": 0.6026, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 0.7431384325027466, | |
| "learning_rate": 2.909090909090909e-05, | |
| "loss": 0.598, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6363636363636362, | |
| "grad_norm": 0.7853900194168091, | |
| "learning_rate": 3.272727272727273e-05, | |
| "loss": 0.6217, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 0.7090263366699219, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 0.5511, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.6501097679138184, | |
| "learning_rate": 4e-05, | |
| "loss": 0.5011, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.5362216830253601, | |
| "eval_runtime": 30.8116, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 0.6643696427345276, | |
| "learning_rate": 4.3636363636363636e-05, | |
| "loss": 0.4868, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.3636363636363638, | |
| "grad_norm": 0.7205197215080261, | |
| "learning_rate": 4.7272727272727275e-05, | |
| "loss": 0.531, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": 0.6881662011146545, | |
| "learning_rate": 5.090909090909091e-05, | |
| "loss": 0.479, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 0.6973153948783875, | |
| "learning_rate": 5.4545454545454546e-05, | |
| "loss": 0.466, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "grad_norm": 0.8126327991485596, | |
| "learning_rate": 5.818181818181818e-05, | |
| "loss": 0.4292, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.48053479194641113, | |
| "eval_runtime": 30.8057, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.090909090909091, | |
| "grad_norm": 0.7639182209968567, | |
| "learning_rate": 6.181818181818182e-05, | |
| "loss": 0.4607, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.2727272727272725, | |
| "grad_norm": 0.627879798412323, | |
| "learning_rate": 6.545454545454546e-05, | |
| "loss": 0.3759, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.4545454545454546, | |
| "grad_norm": 0.7164282202720642, | |
| "learning_rate": 6.90909090909091e-05, | |
| "loss": 0.3921, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 0.6686579585075378, | |
| "learning_rate": 7.272727272727273e-05, | |
| "loss": 0.3881, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.8181818181818183, | |
| "grad_norm": 0.7896100878715515, | |
| "learning_rate": 7.636363636363637e-05, | |
| "loss": 0.3872, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.7360653281211853, | |
| "learning_rate": 8e-05, | |
| "loss": 0.3757, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.46927714347839355, | |
| "eval_runtime": 30.8082, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.181818181818182, | |
| "grad_norm": 0.8323110938072205, | |
| "learning_rate": 8.363636363636364e-05, | |
| "loss": 0.3215, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.363636363636363, | |
| "grad_norm": 0.7733283638954163, | |
| "learning_rate": 8.727272727272727e-05, | |
| "loss": 0.3002, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.545454545454545, | |
| "grad_norm": 0.8626326322555542, | |
| "learning_rate": 9.090909090909092e-05, | |
| "loss": 0.2698, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 1.0941702127456665, | |
| "learning_rate": 9.454545454545455e-05, | |
| "loss": 0.2907, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 4.909090909090909, | |
| "grad_norm": 0.9751712083816528, | |
| "learning_rate": 9.818181818181818e-05, | |
| "loss": 0.2815, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.48815110325813293, | |
| "eval_runtime": 30.8074, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.090909090909091, | |
| "grad_norm": 0.8294689655303955, | |
| "learning_rate": 9.999899300364532e-05, | |
| "loss": 0.2346, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.2727272727272725, | |
| "grad_norm": 0.9779950380325317, | |
| "learning_rate": 9.99909372761763e-05, | |
| "loss": 0.1881, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.454545454545454, | |
| "grad_norm": 1.1621687412261963, | |
| "learning_rate": 9.997482711915927e-05, | |
| "loss": 0.1494, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.636363636363637, | |
| "grad_norm": 1.6494107246398926, | |
| "learning_rate": 9.99506651282272e-05, | |
| "loss": 0.1491, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 5.818181818181818, | |
| "grad_norm": 1.308590054512024, | |
| "learning_rate": 9.991845519630678e-05, | |
| "loss": 0.1526, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.0228348970413208, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.1492, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.5787567496299744, | |
| "eval_runtime": 30.8108, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.181818181818182, | |
| "grad_norm": 0.9201527237892151, | |
| "learning_rate": 9.982991356370404e-05, | |
| "loss": 0.0965, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.363636363636363, | |
| "grad_norm": 0.8013624548912048, | |
| "learning_rate": 9.977359612865423e-05, | |
| "loss": 0.0846, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.545454545454545, | |
| "grad_norm": 1.058500051498413, | |
| "learning_rate": 9.970925928158274e-05, | |
| "loss": 0.0873, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 6.7272727272727275, | |
| "grad_norm": 1.092773199081421, | |
| "learning_rate": 9.963691338830044e-05, | |
| "loss": 0.068, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "grad_norm": 1.0958987474441528, | |
| "learning_rate": 9.955657010501806e-05, | |
| "loss": 0.0722, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.7477759718894958, | |
| "eval_runtime": 30.8103, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 7.090909090909091, | |
| "grad_norm": 1.0380290746688843, | |
| "learning_rate": 9.946824237646824e-05, | |
| "loss": 0.0615, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 7.2727272727272725, | |
| "grad_norm": 1.0997962951660156, | |
| "learning_rate": 9.937194443381972e-05, | |
| "loss": 0.0391, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 7.454545454545454, | |
| "grad_norm": 0.8664355278015137, | |
| "learning_rate": 9.926769179238466e-05, | |
| "loss": 0.0341, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 7.636363636363637, | |
| "grad_norm": 0.9188678860664368, | |
| "learning_rate": 9.915550124911866e-05, | |
| "loss": 0.0459, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 7.818181818181818, | |
| "grad_norm": 1.1214436292648315, | |
| "learning_rate": 9.903539087991462e-05, | |
| "loss": 0.0531, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.0122736692428589, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.0484, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.7403396964073181, | |
| "eval_runtime": 30.8108, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 8.181818181818182, | |
| "grad_norm": 0.6258845329284668, | |
| "learning_rate": 9.877148934427037e-05, | |
| "loss": 0.0277, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 8.363636363636363, | |
| "grad_norm": 0.8622820973396301, | |
| "learning_rate": 9.862774069706346e-05, | |
| "loss": 0.0278, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 8.545454545454545, | |
| "grad_norm": 0.8286793231964111, | |
| "learning_rate": 9.847615725553456e-05, | |
| "loss": 0.0358, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 8.727272727272727, | |
| "grad_norm": 0.6255679130554199, | |
| "learning_rate": 9.831676344247342e-05, | |
| "loss": 0.0259, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 8.909090909090908, | |
| "grad_norm": 0.8187332153320312, | |
| "learning_rate": 9.814958493905963e-05, | |
| "loss": 0.0365, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.8076421618461609, | |
| "eval_runtime": 30.8163, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 9.090909090909092, | |
| "grad_norm": 0.49805107712745667, | |
| "learning_rate": 9.797464868072488e-05, | |
| "loss": 0.029, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 9.272727272727273, | |
| "grad_norm": 0.6919748783111572, | |
| "learning_rate": 9.779198285281325e-05, | |
| "loss": 0.0156, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 9.454545454545455, | |
| "grad_norm": 0.5414004921913147, | |
| "learning_rate": 9.760161688604008e-05, | |
| "loss": 0.021, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 9.636363636363637, | |
| "grad_norm": 0.7409151196479797, | |
| "learning_rate": 9.740358145174998e-05, | |
| "loss": 0.0204, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 9.818181818181818, | |
| "grad_norm": 0.8508411645889282, | |
| "learning_rate": 9.719790845697533e-05, | |
| "loss": 0.0205, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.7725595235824585, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.0191, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.8328536152839661, | |
| "eval_runtime": 30.8136, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 10.181818181818182, | |
| "grad_norm": 0.43200600147247314, | |
| "learning_rate": 9.676378356149734e-05, | |
| "loss": 0.0113, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 10.363636363636363, | |
| "grad_norm": 0.6364129185676575, | |
| "learning_rate": 9.653540160603956e-05, | |
| "loss": 0.0128, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 10.545454545454545, | |
| "grad_norm": 0.7452590465545654, | |
| "learning_rate": 9.629952196931901e-05, | |
| "loss": 0.0129, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 10.727272727272727, | |
| "grad_norm": 0.689669132232666, | |
| "learning_rate": 9.60561826557425e-05, | |
| "loss": 0.0192, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "grad_norm": 0.6580386757850647, | |
| "learning_rate": 9.580542287160348e-05, | |
| "loss": 0.0134, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.8433478474617004, | |
| "eval_runtime": 30.8076, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "step": 121, | |
| "total_flos": 1.1820445555123814e+17, | |
| "train_loss": 0.25857353995531057, | |
| "train_runtime": 4284.2126, | |
| "train_samples_per_second": 1.027, | |
| "train_steps_per_second": 0.128 | |
| } | |
| ], | |
| "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": 1.1820445555123814e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |