| { | |
| "best_metric": 0.767100989818573, | |
| "best_model_checkpoint": "data/Mistral-7B_task-2_120-samples_config-1_full_auto/checkpoint-55", | |
| "epoch": 12.0, | |
| "eval_steps": 500, | |
| "global_step": 132, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09090909090909091, | |
| "grad_norm": 1.0057252645492554, | |
| "learning_rate": 1.818181818181818e-06, | |
| "loss": 1.0806, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": 1.1914880275726318, | |
| "learning_rate": 3.636363636363636e-06, | |
| "loss": 1.216, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 0.9636508822441101, | |
| "learning_rate": 7.272727272727272e-06, | |
| "loss": 1.109, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.5454545454545454, | |
| "grad_norm": 0.9821034669876099, | |
| "learning_rate": 1.0909090909090909e-05, | |
| "loss": 1.1174, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.8801716566085815, | |
| "learning_rate": 1.4545454545454545e-05, | |
| "loss": 1.1537, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": 0.6772984862327576, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 1.067, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.0422520637512207, | |
| "eval_runtime": 30.7848, | |
| "eval_samples_per_second": 0.78, | |
| "eval_steps_per_second": 0.78, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 0.627533495426178, | |
| "learning_rate": 2.1818181818181818e-05, | |
| "loss": 1.0791, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.2727272727272727, | |
| "grad_norm": 0.39374202489852905, | |
| "learning_rate": 2.5454545454545454e-05, | |
| "loss": 0.998, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 0.4058741629123688, | |
| "learning_rate": 2.909090909090909e-05, | |
| "loss": 0.997, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6363636363636362, | |
| "grad_norm": 0.40029197931289673, | |
| "learning_rate": 3.272727272727273e-05, | |
| "loss": 1.016, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 0.4022189676761627, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 0.9591, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.39615121483802795, | |
| "learning_rate": 4e-05, | |
| "loss": 0.8773, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.9271183013916016, | |
| "eval_runtime": 30.7876, | |
| "eval_samples_per_second": 0.78, | |
| "eval_steps_per_second": 0.78, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 0.3347851037979126, | |
| "learning_rate": 4.3636363636363636e-05, | |
| "loss": 0.9102, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.3636363636363638, | |
| "grad_norm": 0.486189067363739, | |
| "learning_rate": 4.7272727272727275e-05, | |
| "loss": 0.8838, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": 0.4024110436439514, | |
| "learning_rate": 5.090909090909091e-05, | |
| "loss": 0.8295, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 0.34966298937797546, | |
| "learning_rate": 5.4545454545454546e-05, | |
| "loss": 0.845, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "grad_norm": 0.3345252275466919, | |
| "learning_rate": 5.818181818181818e-05, | |
| "loss": 0.781, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.8065378069877625, | |
| "eval_runtime": 30.7877, | |
| "eval_samples_per_second": 0.78, | |
| "eval_steps_per_second": 0.78, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.090909090909091, | |
| "grad_norm": 0.35362398624420166, | |
| "learning_rate": 6.181818181818182e-05, | |
| "loss": 0.7874, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.2727272727272725, | |
| "grad_norm": 0.2822090685367584, | |
| "learning_rate": 6.545454545454546e-05, | |
| "loss": 0.7516, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.4545454545454546, | |
| "grad_norm": 0.25855928659439087, | |
| "learning_rate": 6.90909090909091e-05, | |
| "loss": 0.7592, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 0.24022454023361206, | |
| "learning_rate": 7.272727272727273e-05, | |
| "loss": 0.7148, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.8181818181818183, | |
| "grad_norm": 0.33709412813186646, | |
| "learning_rate": 7.636363636363637e-05, | |
| "loss": 0.6904, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.2493952363729477, | |
| "learning_rate": 8e-05, | |
| "loss": 0.79, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.7789912223815918, | |
| "eval_runtime": 30.7917, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.181818181818182, | |
| "grad_norm": 0.28979021310806274, | |
| "learning_rate": 8.363636363636364e-05, | |
| "loss": 0.6887, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.363636363636363, | |
| "grad_norm": 0.2719300091266632, | |
| "learning_rate": 8.727272727272727e-05, | |
| "loss": 0.7311, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.545454545454545, | |
| "grad_norm": 0.2569809854030609, | |
| "learning_rate": 9.090909090909092e-05, | |
| "loss": 0.693, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 0.3406239151954651, | |
| "learning_rate": 9.454545454545455e-05, | |
| "loss": 0.6996, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 4.909090909090909, | |
| "grad_norm": 0.2898145318031311, | |
| "learning_rate": 9.818181818181818e-05, | |
| "loss": 0.7295, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.767100989818573, | |
| "eval_runtime": 30.7931, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.090909090909091, | |
| "grad_norm": 0.27576756477355957, | |
| "learning_rate": 9.999899300364532e-05, | |
| "loss": 0.6792, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.2727272727272725, | |
| "grad_norm": 0.31847649812698364, | |
| "learning_rate": 9.99909372761763e-05, | |
| "loss": 0.6869, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.454545454545454, | |
| "grad_norm": 0.32753080129623413, | |
| "learning_rate": 9.997482711915927e-05, | |
| "loss": 0.624, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.636363636363637, | |
| "grad_norm": 0.3538777828216553, | |
| "learning_rate": 9.99506651282272e-05, | |
| "loss": 0.6434, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 5.818181818181818, | |
| "grad_norm": 0.36934995651245117, | |
| "learning_rate": 9.991845519630678e-05, | |
| "loss": 0.6453, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.3215658962726593, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.6803, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.7683818936347961, | |
| "eval_runtime": 30.795, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.181818181818182, | |
| "grad_norm": 0.41097292304039, | |
| "learning_rate": 9.982991356370404e-05, | |
| "loss": 0.5753, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.363636363636363, | |
| "grad_norm": 0.41823679208755493, | |
| "learning_rate": 9.977359612865423e-05, | |
| "loss": 0.611, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.545454545454545, | |
| "grad_norm": 0.4454464018344879, | |
| "learning_rate": 9.970925928158274e-05, | |
| "loss": 0.5511, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 6.7272727272727275, | |
| "grad_norm": 0.462645560503006, | |
| "learning_rate": 9.963691338830044e-05, | |
| "loss": 0.5894, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "grad_norm": 0.4251982569694519, | |
| "learning_rate": 9.955657010501806e-05, | |
| "loss": 0.6231, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.7836182713508606, | |
| "eval_runtime": 30.7954, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 7.090909090909091, | |
| "grad_norm": 0.4386478066444397, | |
| "learning_rate": 9.946824237646824e-05, | |
| "loss": 0.6288, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 7.2727272727272725, | |
| "grad_norm": 0.4783054292201996, | |
| "learning_rate": 9.937194443381972e-05, | |
| "loss": 0.476, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 7.454545454545454, | |
| "grad_norm": 0.7922077178955078, | |
| "learning_rate": 9.926769179238466e-05, | |
| "loss": 0.4917, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 7.636363636363637, | |
| "grad_norm": 0.5860278010368347, | |
| "learning_rate": 9.915550124911866e-05, | |
| "loss": 0.5216, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 7.818181818181818, | |
| "grad_norm": 0.5895577669143677, | |
| "learning_rate": 9.903539087991462e-05, | |
| "loss": 0.5065, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.6083407402038574, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.5575, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.8162687420845032, | |
| "eval_runtime": 30.7879, | |
| "eval_samples_per_second": 0.78, | |
| "eval_steps_per_second": 0.78, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 8.181818181818182, | |
| "grad_norm": 0.6505959033966064, | |
| "learning_rate": 9.877148934427037e-05, | |
| "loss": 0.4805, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 8.363636363636363, | |
| "grad_norm": 0.9110755920410156, | |
| "learning_rate": 9.862774069706346e-05, | |
| "loss": 0.4371, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 8.545454545454545, | |
| "grad_norm": 0.6465612053871155, | |
| "learning_rate": 9.847615725553456e-05, | |
| "loss": 0.4602, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 8.727272727272727, | |
| "grad_norm": 0.6882513165473938, | |
| "learning_rate": 9.831676344247342e-05, | |
| "loss": 0.3858, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 8.909090909090908, | |
| "grad_norm": 0.7937643527984619, | |
| "learning_rate": 9.814958493905963e-05, | |
| "loss": 0.452, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.8697136044502258, | |
| "eval_runtime": 30.7907, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 9.090909090909092, | |
| "grad_norm": 0.7437408566474915, | |
| "learning_rate": 9.797464868072488e-05, | |
| "loss": 0.4428, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 9.272727272727273, | |
| "grad_norm": 0.938471794128418, | |
| "learning_rate": 9.779198285281325e-05, | |
| "loss": 0.3484, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 9.454545454545455, | |
| "grad_norm": 0.8467069864273071, | |
| "learning_rate": 9.760161688604008e-05, | |
| "loss": 0.3539, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 9.636363636363637, | |
| "grad_norm": 1.062785029411316, | |
| "learning_rate": 9.740358145174998e-05, | |
| "loss": 0.408, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 9.818181818181818, | |
| "grad_norm": 1.0526679754257202, | |
| "learning_rate": 9.719790845697533e-05, | |
| "loss": 0.37, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.1780627965927124, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.3535, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.904864490032196, | |
| "eval_runtime": 30.7921, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 10.181818181818182, | |
| "grad_norm": 0.8502280116081238, | |
| "learning_rate": 9.676378356149734e-05, | |
| "loss": 0.2876, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 10.363636363636363, | |
| "grad_norm": 0.975743293762207, | |
| "learning_rate": 9.653540160603956e-05, | |
| "loss": 0.2938, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 10.545454545454545, | |
| "grad_norm": 0.8724481463432312, | |
| "learning_rate": 9.629952196931901e-05, | |
| "loss": 0.291, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 10.727272727272727, | |
| "grad_norm": 0.8257167935371399, | |
| "learning_rate": 9.60561826557425e-05, | |
| "loss": 0.3074, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "grad_norm": 0.970210611820221, | |
| "learning_rate": 9.580542287160348e-05, | |
| "loss": 0.3253, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.9732216000556946, | |
| "eval_runtime": 30.7947, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 11.090909090909092, | |
| "grad_norm": 0.8243751525878906, | |
| "learning_rate": 9.554728301876526e-05, | |
| "loss": 0.2698, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 11.272727272727273, | |
| "grad_norm": 0.8600305914878845, | |
| "learning_rate": 9.528180468815155e-05, | |
| "loss": 0.2341, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 11.454545454545455, | |
| "grad_norm": 0.915551483631134, | |
| "learning_rate": 9.50090306530454e-05, | |
| "loss": 0.2473, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 11.636363636363637, | |
| "grad_norm": 0.8568307161331177, | |
| "learning_rate": 9.472900486219769e-05, | |
| "loss": 0.2733, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 11.818181818181818, | |
| "grad_norm": 1.049709677696228, | |
| "learning_rate": 9.444177243274618e-05, | |
| "loss": 0.2352, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.0510989427566528, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.2393, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.0469673871994019, | |
| "eval_runtime": 30.7885, | |
| "eval_samples_per_second": 0.78, | |
| "eval_steps_per_second": 0.78, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "step": 132, | |
| "total_flos": 1.2895031521391411e+17, | |
| "train_loss": 0.6271772569779194, | |
| "train_runtime": 4671.7627, | |
| "train_samples_per_second": 0.942, | |
| "train_steps_per_second": 0.118 | |
| } | |
| ], | |
| "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": 1.2895031521391411e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |