| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.030214903501151944, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0006042980700230389, | |
| "grad_norm": 0.5589237809181213, | |
| "learning_rate": 1e-05, | |
| "loss": 3.1692, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0006042980700230389, | |
| "eval_loss": 3.1913259029388428, | |
| "eval_runtime": 229.6838, | |
| "eval_samples_per_second": 12.134, | |
| "eval_steps_per_second": 6.069, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0012085961400460777, | |
| "grad_norm": 0.4362795948982239, | |
| "learning_rate": 2e-05, | |
| "loss": 2.8003, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0018128942100691166, | |
| "grad_norm": 0.4628395438194275, | |
| "learning_rate": 3e-05, | |
| "loss": 3.3741, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0024171922800921555, | |
| "grad_norm": 0.4753565788269043, | |
| "learning_rate": 4e-05, | |
| "loss": 2.9939, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.003021490350115194, | |
| "grad_norm": 0.5600993037223816, | |
| "learning_rate": 5e-05, | |
| "loss": 3.2337, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0036257884201382332, | |
| "grad_norm": 0.4630330801010132, | |
| "learning_rate": 6e-05, | |
| "loss": 2.8722, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.004230086490161272, | |
| "grad_norm": 0.551735520362854, | |
| "learning_rate": 7e-05, | |
| "loss": 3.0785, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.004834384560184311, | |
| "grad_norm": 0.5148279666900635, | |
| "learning_rate": 8e-05, | |
| "loss": 2.8164, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.00543868263020735, | |
| "grad_norm": 0.7112084627151489, | |
| "learning_rate": 9e-05, | |
| "loss": 3.3676, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.006042980700230388, | |
| "grad_norm": 0.5696298480033875, | |
| "learning_rate": 0.0001, | |
| "loss": 2.7856, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.006647278770253427, | |
| "grad_norm": 0.6853129863739014, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 3.1526, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0072515768402764665, | |
| "grad_norm": 0.6386879086494446, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 2.706, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.007855874910299505, | |
| "grad_norm": 0.7042403817176819, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 2.5668, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.007855874910299505, | |
| "eval_loss": 2.8041810989379883, | |
| "eval_runtime": 78.0564, | |
| "eval_samples_per_second": 35.705, | |
| "eval_steps_per_second": 17.859, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.008460172980322544, | |
| "grad_norm": 0.7668046355247498, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 2.9344, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.009064471050345583, | |
| "grad_norm": 0.8007228374481201, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 2.8764, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.009668769120368622, | |
| "grad_norm": 0.6990913152694702, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 2.4552, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.010273067190391661, | |
| "grad_norm": 0.6952396035194397, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 2.5646, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0108773652604147, | |
| "grad_norm": 0.7194958925247192, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 2.2515, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.01148166333043774, | |
| "grad_norm": 0.7544381618499756, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 2.5523, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.012085961400460777, | |
| "grad_norm": 0.7936828136444092, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 2.4444, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.012690259470483816, | |
| "grad_norm": 0.7764612436294556, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 2.3171, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.013294557540506855, | |
| "grad_norm": 0.9252168536186218, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 2.5806, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.013898855610529894, | |
| "grad_norm": 0.8082977533340454, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 2.2255, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.014503153680552933, | |
| "grad_norm": 0.8068621158599854, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 2.1052, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.015107451750575972, | |
| "grad_norm": 0.876987874507904, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 2.0828, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01571174982059901, | |
| "grad_norm": 0.9265044331550598, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 2.1888, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.01571174982059901, | |
| "eval_loss": 2.1405892372131348, | |
| "eval_runtime": 77.9878, | |
| "eval_samples_per_second": 35.736, | |
| "eval_steps_per_second": 17.875, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.01631604789062205, | |
| "grad_norm": 0.8348265290260315, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 2.0727, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.016920345960645088, | |
| "grad_norm": 0.8782357573509216, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 2.1957, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.017524644030668127, | |
| "grad_norm": 0.8670806884765625, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 1.9359, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.018128942100691166, | |
| "grad_norm": 0.9028540849685669, | |
| "learning_rate": 5e-05, | |
| "loss": 1.9664, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.018733240170714205, | |
| "grad_norm": 0.8907434940338135, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 1.969, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.019337538240737244, | |
| "grad_norm": 0.8233531713485718, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 1.974, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.019941836310760283, | |
| "grad_norm": 0.8587433099746704, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 1.8521, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.020546134380783322, | |
| "grad_norm": 0.897334098815918, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 1.9035, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.02115043245080636, | |
| "grad_norm": 0.9631185531616211, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 1.9148, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0217547305208294, | |
| "grad_norm": 1.0168321132659912, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 1.9873, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.02235902859085244, | |
| "grad_norm": 0.8644974827766418, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 1.8347, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.02296332666087548, | |
| "grad_norm": 0.8866729140281677, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 2.0322, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.023567624730898514, | |
| "grad_norm": 0.9013758897781372, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 1.8802, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.023567624730898514, | |
| "eval_loss": 1.8680973052978516, | |
| "eval_runtime": 78.0813, | |
| "eval_samples_per_second": 35.694, | |
| "eval_steps_per_second": 17.853, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.024171922800921553, | |
| "grad_norm": 0.877164363861084, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 1.7625, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.024776220870944592, | |
| "grad_norm": 0.9773980975151062, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 1.9383, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.02538051894096763, | |
| "grad_norm": 0.8631694316864014, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 1.8411, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.02598481701099067, | |
| "grad_norm": 0.8837679624557495, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 1.8574, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.02658911508101371, | |
| "grad_norm": 0.9769808650016785, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 1.9396, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.02719341315103675, | |
| "grad_norm": 0.948658287525177, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 2.0352, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.027797711221059788, | |
| "grad_norm": 0.9376592040061951, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 1.7759, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.028402009291082827, | |
| "grad_norm": 0.81429523229599, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 1.9055, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.029006307361105866, | |
| "grad_norm": 0.9344580769538879, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 2.015, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.029610605431128905, | |
| "grad_norm": 0.8902539014816284, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 1.7384, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.030214903501151944, | |
| "grad_norm": 0.9016924500465393, | |
| "learning_rate": 0.0, | |
| "loss": 1.6748, | |
| "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": 4.509070769258496e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |