| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.024783147459727387, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0004956629491945477, | |
| "grad_norm": 0.7391602396965027, | |
| "learning_rate": 1e-05, | |
| "loss": 2.3969, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0004956629491945477, | |
| "eval_loss": 2.63633394241333, | |
| "eval_runtime": 19.1271, | |
| "eval_samples_per_second": 88.827, | |
| "eval_steps_per_second": 11.136, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0009913258983890955, | |
| "grad_norm": 0.8179469704627991, | |
| "learning_rate": 2e-05, | |
| "loss": 2.3856, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.001486988847583643, | |
| "grad_norm": 0.7126672863960266, | |
| "learning_rate": 3e-05, | |
| "loss": 2.3, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.001982651796778191, | |
| "grad_norm": 1.0962400436401367, | |
| "learning_rate": 4e-05, | |
| "loss": 3.1446, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0024783147459727386, | |
| "grad_norm": 0.7755838632583618, | |
| "learning_rate": 5e-05, | |
| "loss": 2.4778, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0024783147459727386, | |
| "eval_loss": 2.6211066246032715, | |
| "eval_runtime": 17.2492, | |
| "eval_samples_per_second": 98.497, | |
| "eval_steps_per_second": 12.348, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.002973977695167286, | |
| "grad_norm": 1.0329855680465698, | |
| "learning_rate": 6e-05, | |
| "loss": 3.06, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.003469640644361834, | |
| "grad_norm": 0.7987164855003357, | |
| "learning_rate": 7e-05, | |
| "loss": 2.5097, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.003965303593556382, | |
| "grad_norm": 0.8003743886947632, | |
| "learning_rate": 8e-05, | |
| "loss": 2.4667, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0044609665427509295, | |
| "grad_norm": 0.7125903367996216, | |
| "learning_rate": 9e-05, | |
| "loss": 2.2826, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.004956629491945477, | |
| "grad_norm": 0.8778437376022339, | |
| "learning_rate": 0.0001, | |
| "loss": 2.7076, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.004956629491945477, | |
| "eval_loss": 2.435589075088501, | |
| "eval_runtime": 17.2728, | |
| "eval_samples_per_second": 98.363, | |
| "eval_steps_per_second": 12.332, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.005452292441140025, | |
| "grad_norm": 0.7835959196090698, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 2.587, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.005947955390334572, | |
| "grad_norm": 0.6849549412727356, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 2.4522, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.00644361833952912, | |
| "grad_norm": 0.7166538238525391, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 2.4591, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.006939281288723668, | |
| "grad_norm": 0.5679525136947632, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 2.0319, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.007434944237918215, | |
| "grad_norm": 0.6390686631202698, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 2.2686, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.007434944237918215, | |
| "eval_loss": 2.1834659576416016, | |
| "eval_runtime": 17.2849, | |
| "eval_samples_per_second": 98.294, | |
| "eval_steps_per_second": 12.323, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.007930607187112764, | |
| "grad_norm": 0.6615875959396362, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 2.2409, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.00842627013630731, | |
| "grad_norm": 0.5758230686187744, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 1.8906, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.008921933085501859, | |
| "grad_norm": 0.6313810348510742, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 2.0695, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.009417596034696406, | |
| "grad_norm": 0.4957845211029053, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 1.7364, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.009913258983890954, | |
| "grad_norm": 0.6371930837631226, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 2.0556, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.009913258983890954, | |
| "eval_loss": 1.9821568727493286, | |
| "eval_runtime": 17.2924, | |
| "eval_samples_per_second": 98.251, | |
| "eval_steps_per_second": 12.318, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.010408921933085501, | |
| "grad_norm": 0.6001903414726257, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 1.9351, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.01090458488228005, | |
| "grad_norm": 0.654202401638031, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 1.889, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.011400247831474598, | |
| "grad_norm": 0.6593927145004272, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 2.0221, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.011895910780669145, | |
| "grad_norm": 0.48587918281555176, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 1.5167, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.012391573729863693, | |
| "grad_norm": 0.5013470649719238, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 1.581, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.012391573729863693, | |
| "eval_loss": 1.8297536373138428, | |
| "eval_runtime": 17.273, | |
| "eval_samples_per_second": 98.361, | |
| "eval_steps_per_second": 12.331, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01288723667905824, | |
| "grad_norm": 0.7884762287139893, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 2.0006, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.013382899628252789, | |
| "grad_norm": 0.6535519957542419, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 1.6955, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.013878562577447335, | |
| "grad_norm": 0.6614362597465515, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 1.7235, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.014374225526641884, | |
| "grad_norm": 0.569136917591095, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 1.6187, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.01486988847583643, | |
| "grad_norm": 0.8128046989440918, | |
| "learning_rate": 5e-05, | |
| "loss": 1.8391, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.01486988847583643, | |
| "eval_loss": 1.7055069208145142, | |
| "eval_runtime": 17.3629, | |
| "eval_samples_per_second": 97.852, | |
| "eval_steps_per_second": 12.268, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.015365551425030979, | |
| "grad_norm": 0.5898133516311646, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 1.4803, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.015861214374225528, | |
| "grad_norm": 0.8634248375892639, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 1.9185, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.016356877323420074, | |
| "grad_norm": 0.7606371641159058, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 1.7444, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.01685254027261462, | |
| "grad_norm": 0.827926516532898, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 1.5359, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.01734820322180917, | |
| "grad_norm": 0.6055005192756653, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 1.3024, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.01734820322180917, | |
| "eval_loss": 1.6104322671890259, | |
| "eval_runtime": 17.2589, | |
| "eval_samples_per_second": 98.442, | |
| "eval_steps_per_second": 12.341, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.017843866171003718, | |
| "grad_norm": 0.8044369220733643, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 1.5692, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.018339529120198265, | |
| "grad_norm": 0.6717920303344727, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 1.6285, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.01883519206939281, | |
| "grad_norm": 0.7943913340568542, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 1.5588, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.019330855018587362, | |
| "grad_norm": 0.7974594235420227, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 1.4334, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.01982651796778191, | |
| "grad_norm": 0.7847771048545837, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 1.6791, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.01982651796778191, | |
| "eval_loss": 1.5548481941223145, | |
| "eval_runtime": 17.2904, | |
| "eval_samples_per_second": 98.263, | |
| "eval_steps_per_second": 12.319, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.020322180916976455, | |
| "grad_norm": 0.6804707050323486, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 1.5037, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.020817843866171002, | |
| "grad_norm": 0.8448418378829956, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 1.6357, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.021313506815365552, | |
| "grad_norm": 0.8616818189620972, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 1.5866, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.0218091697645601, | |
| "grad_norm": 0.6975346803665161, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 1.4729, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.022304832713754646, | |
| "grad_norm": 0.6911585927009583, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 1.3803, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.022304832713754646, | |
| "eval_loss": 1.5319888591766357, | |
| "eval_runtime": 17.2847, | |
| "eval_samples_per_second": 98.295, | |
| "eval_steps_per_second": 12.323, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.022800495662949196, | |
| "grad_norm": 0.6003568172454834, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 1.4585, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.023296158612143743, | |
| "grad_norm": 0.7089613676071167, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 1.4629, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.02379182156133829, | |
| "grad_norm": 0.7499310970306396, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 1.4805, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.024287484510532836, | |
| "grad_norm": 0.7139114141464233, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 1.5951, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.024783147459727387, | |
| "grad_norm": 0.7746477127075195, | |
| "learning_rate": 0.0, | |
| "loss": 1.6068, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.024783147459727387, | |
| "eval_loss": 1.527747392654419, | |
| "eval_runtime": 17.2729, | |
| "eval_samples_per_second": 98.362, | |
| "eval_steps_per_second": 12.331, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 763606723461120.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |