| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.005094113751560072, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00010188227503120145, | |
| "grad_norm": 14.901717185974121, | |
| "learning_rate": 1e-05, | |
| "loss": 3.63, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00010188227503120145, | |
| "eval_loss": 3.6769039630889893, | |
| "eval_runtime": 654.3291, | |
| "eval_samples_per_second": 25.264, | |
| "eval_steps_per_second": 12.633, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0002037645500624029, | |
| "grad_norm": 13.94775104522705, | |
| "learning_rate": 2e-05, | |
| "loss": 3.7484, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00030564682509360435, | |
| "grad_norm": 14.099560737609863, | |
| "learning_rate": 3e-05, | |
| "loss": 3.4832, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0004075291001248058, | |
| "grad_norm": 14.928464889526367, | |
| "learning_rate": 4e-05, | |
| "loss": 3.1936, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0005094113751560073, | |
| "grad_norm": 14.232325553894043, | |
| "learning_rate": 5e-05, | |
| "loss": 2.7632, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0006112936501872087, | |
| "grad_norm": 13.29269790649414, | |
| "learning_rate": 6e-05, | |
| "loss": 2.4072, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0007131759252184102, | |
| "grad_norm": 14.143783569335938, | |
| "learning_rate": 7e-05, | |
| "loss": 2.1272, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0008150582002496116, | |
| "grad_norm": 12.234458923339844, | |
| "learning_rate": 8e-05, | |
| "loss": 1.9853, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0009169404752808131, | |
| "grad_norm": 11.770651817321777, | |
| "learning_rate": 9e-05, | |
| "loss": 1.2758, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0010188227503120145, | |
| "grad_norm": 10.147363662719727, | |
| "learning_rate": 0.0001, | |
| "loss": 0.8417, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0011207050253432159, | |
| "grad_norm": 9.32343578338623, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 0.6388, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0012225873003744174, | |
| "grad_norm": 6.439914703369141, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 0.37, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0013244695754056188, | |
| "grad_norm": 1.927167296409607, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 0.2115, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0013244695754056188, | |
| "eval_loss": 0.2436702996492386, | |
| "eval_runtime": 578.2604, | |
| "eval_samples_per_second": 28.587, | |
| "eval_steps_per_second": 14.295, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0014263518504368203, | |
| "grad_norm": 3.150268077850342, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.4769, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0015282341254680217, | |
| "grad_norm": 11.413074493408203, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.1393, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0016301164004992232, | |
| "grad_norm": 2.8546557426452637, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 0.2937, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0017319986755304246, | |
| "grad_norm": 1.8057975769042969, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 0.1379, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0018338809505616261, | |
| "grad_norm": 1.8448413610458374, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.204, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0019357632255928275, | |
| "grad_norm": 11.360038757324219, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 0.5119, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.002037645500624029, | |
| "grad_norm": 6.971611022949219, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.0774, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0021395277756552306, | |
| "grad_norm": 0.6716346740722656, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 0.0313, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0022414100506864317, | |
| "grad_norm": 1.3120988607406616, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.1035, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0023432923257176333, | |
| "grad_norm": 1.7842707633972168, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 0.0699, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.002445174600748835, | |
| "grad_norm": 0.9044802784919739, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 0.1219, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.002547056875780036, | |
| "grad_norm": 0.16728538274765015, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.0307, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0026489391508112375, | |
| "grad_norm": 0.81120365858078, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.0296, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0026489391508112375, | |
| "eval_loss": 0.07582408934831619, | |
| "eval_runtime": 578.2782, | |
| "eval_samples_per_second": 28.587, | |
| "eval_steps_per_second": 14.294, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.002750821425842439, | |
| "grad_norm": 2.156219720840454, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 0.1494, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0028527037008736406, | |
| "grad_norm": 0.3473513722419739, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 0.0356, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0029545859759048418, | |
| "grad_norm": 1.3672902584075928, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 0.1603, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0030564682509360433, | |
| "grad_norm": 2.2377445697784424, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0971, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.003158350525967245, | |
| "grad_norm": 0.6241983771324158, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 0.042, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.0032602328009984465, | |
| "grad_norm": 0.8559085130691528, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 0.0415, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.0033621150760296476, | |
| "grad_norm": 0.60133296251297, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 0.0377, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.003463997351060849, | |
| "grad_norm": 0.6471958160400391, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.0148, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.0035658796260920507, | |
| "grad_norm": 0.7025775909423828, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.0149, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0036677619011232523, | |
| "grad_norm": 0.6499977707862854, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.0519, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0037696441761544534, | |
| "grad_norm": 0.6970950365066528, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 0.0287, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.003871526451185655, | |
| "grad_norm": 0.49171292781829834, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.0359, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.003973408726216856, | |
| "grad_norm": 0.8748102188110352, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 0.0462, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.003973408726216856, | |
| "eval_loss": 0.05458526685833931, | |
| "eval_runtime": 577.9867, | |
| "eval_samples_per_second": 28.601, | |
| "eval_steps_per_second": 14.301, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.004075291001248058, | |
| "grad_norm": 2.2403550148010254, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.1941, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.004177173276279259, | |
| "grad_norm": 1.1056747436523438, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 0.1054, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.004279055551310461, | |
| "grad_norm": 0.17147065699100494, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0343, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.004380937826341662, | |
| "grad_norm": 0.2568279802799225, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 0.0375, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.0044828201013728634, | |
| "grad_norm": 0.33789515495300293, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 0.0438, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.004584702376404065, | |
| "grad_norm": 0.9882907271385193, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0627, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0046865846514352666, | |
| "grad_norm": 0.8271799087524414, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.0217, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.004788466926466468, | |
| "grad_norm": 0.8588903546333313, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 0.0656, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.00489034920149767, | |
| "grad_norm": 0.18504083156585693, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.0376, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.004992231476528871, | |
| "grad_norm": 0.46786996722221375, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 0.0398, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.005094113751560072, | |
| "grad_norm": 0.7483496069908142, | |
| "learning_rate": 0.0, | |
| "loss": 0.068, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.4419086016512e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |