| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.991150442477876, | |
| "eval_steps": 500, | |
| "global_step": 252, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.058997050147492625, | |
| "grad_norm": 2.2323501110076904, | |
| "learning_rate": 1.923076923076923e-05, | |
| "loss": 0.9779, | |
| "mean_token_accuracy": 0.7360278993844986, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.11799410029498525, | |
| "grad_norm": 0.8943867087364197, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.8642, | |
| "mean_token_accuracy": 0.7512260988354683, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.17699115044247787, | |
| "grad_norm": 0.6406371593475342, | |
| "learning_rate": 4.9992225160291885e-05, | |
| "loss": 0.7756, | |
| "mean_token_accuracy": 0.7676111996173859, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2359882005899705, | |
| "grad_norm": 0.47420650720596313, | |
| "learning_rate": 4.990481990931694e-05, | |
| "loss": 0.74, | |
| "mean_token_accuracy": 0.7747218057513237, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2949852507374631, | |
| "grad_norm": 0.41185876727104187, | |
| "learning_rate": 4.9720669537317554e-05, | |
| "loss": 0.7214, | |
| "mean_token_accuracy": 0.7785332903265954, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.35398230088495575, | |
| "grad_norm": 0.377851277589798, | |
| "learning_rate": 4.944056921471231e-05, | |
| "loss": 0.7044, | |
| "mean_token_accuracy": 0.7821874514222145, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.41297935103244837, | |
| "grad_norm": 0.3450012803077698, | |
| "learning_rate": 4.9065728428554405e-05, | |
| "loss": 0.6841, | |
| "mean_token_accuracy": 0.7874245882034302, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.471976401179941, | |
| "grad_norm": 0.3287171423435211, | |
| "learning_rate": 4.859776575990668e-05, | |
| "loss": 0.6713, | |
| "mean_token_accuracy": 0.7902060091495514, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5309734513274337, | |
| "grad_norm": 0.3226357698440552, | |
| "learning_rate": 4.8038701894728526e-05, | |
| "loss": 0.6553, | |
| "mean_token_accuracy": 0.7945271402597427, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5899705014749262, | |
| "grad_norm": 0.3185617923736572, | |
| "learning_rate": 4.7390950898453896e-05, | |
| "loss": 0.6583, | |
| "mean_token_accuracy": 0.7935813069343567, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6489675516224189, | |
| "grad_norm": 0.33868589997291565, | |
| "learning_rate": 4.665730979193731e-05, | |
| "loss": 0.6574, | |
| "mean_token_accuracy": 0.793527738749981, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.7079646017699115, | |
| "grad_norm": 0.3180258274078369, | |
| "learning_rate": 4.584094647377942e-05, | |
| "loss": 0.6508, | |
| "mean_token_accuracy": 0.7950170502066612, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7669616519174042, | |
| "grad_norm": 0.37045028805732727, | |
| "learning_rate": 4.494538604118408e-05, | |
| "loss": 0.649, | |
| "mean_token_accuracy": 0.7954698503017426, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.8259587020648967, | |
| "grad_norm": 0.3652035593986511, | |
| "learning_rate": 4.3974495568414175e-05, | |
| "loss": 0.6488, | |
| "mean_token_accuracy": 0.7957066431641578, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8849557522123894, | |
| "grad_norm": 0.33792272210121155, | |
| "learning_rate": 4.293246740857364e-05, | |
| "loss": 0.6398, | |
| "mean_token_accuracy": 0.7980766043066978, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.943952802359882, | |
| "grad_norm": 0.31773921847343445, | |
| "learning_rate": 4.182380109081926e-05, | |
| "loss": 0.6406, | |
| "mean_token_accuracy": 0.7975733190774917, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.0117994100294985, | |
| "grad_norm": 0.7647905945777893, | |
| "learning_rate": 4.065328389117115e-05, | |
| "loss": 0.7546, | |
| "mean_token_accuracy": 0.8012310644475426, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.0707964601769913, | |
| "grad_norm": 0.4024263620376587, | |
| "learning_rate": 3.9425970160818e-05, | |
| "loss": 0.5884, | |
| "mean_token_accuracy": 0.8105781555175782, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.1297935103244838, | |
| "grad_norm": 0.31481897830963135, | |
| "learning_rate": 3.814715950117856e-05, | |
| "loss": 0.594, | |
| "mean_token_accuracy": 0.8084972694516182, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.1887905604719764, | |
| "grad_norm": 0.3306266963481903, | |
| "learning_rate": 3.682237387996043e-05, | |
| "loss": 0.5868, | |
| "mean_token_accuracy": 0.8107855960726738, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.247787610619469, | |
| "grad_norm": 0.3450247347354889, | |
| "learning_rate": 3.545733378703032e-05, | |
| "loss": 0.5885, | |
| "mean_token_accuracy": 0.8103181079030037, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.3067846607669615, | |
| "grad_norm": 0.2881079912185669, | |
| "learning_rate": 3.405793353305565e-05, | |
| "loss": 0.5855, | |
| "mean_token_accuracy": 0.8110026270151138, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.3657817109144543, | |
| "grad_norm": 0.3721728026866913, | |
| "learning_rate": 3.2630215797579524e-05, | |
| "loss": 0.5918, | |
| "mean_token_accuracy": 0.8090571373701095, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.424778761061947, | |
| "grad_norm": 0.28813228011131287, | |
| "learning_rate": 3.1180345536431366e-05, | |
| "loss": 0.5882, | |
| "mean_token_accuracy": 0.8099654763936996, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.4837758112094395, | |
| "grad_norm": 0.32398104667663574, | |
| "learning_rate": 2.971458336114263e-05, | |
| "loss": 0.5791, | |
| "mean_token_accuracy": 0.8125564515590668, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.5427728613569323, | |
| "grad_norm": 0.3099663555622101, | |
| "learning_rate": 2.8239258505316425e-05, | |
| "loss": 0.5852, | |
| "mean_token_accuracy": 0.8109115019440651, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.6017699115044248, | |
| "grad_norm": 0.328464150428772, | |
| "learning_rate": 2.6760741494683584e-05, | |
| "loss": 0.5745, | |
| "mean_token_accuracy": 0.814112713932991, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.6607669616519174, | |
| "grad_norm": 0.35260599851608276, | |
| "learning_rate": 2.5285416638857372e-05, | |
| "loss": 0.575, | |
| "mean_token_accuracy": 0.8141456112265587, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.71976401179941, | |
| "grad_norm": 0.31594163179397583, | |
| "learning_rate": 2.3819654463568643e-05, | |
| "loss": 0.5739, | |
| "mean_token_accuracy": 0.8143744096159935, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 1.7787610619469025, | |
| "grad_norm": 0.28481170535087585, | |
| "learning_rate": 2.2369784202420478e-05, | |
| "loss": 0.5774, | |
| "mean_token_accuracy": 0.8129270792007446, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.8377581120943953, | |
| "grad_norm": 0.25970932841300964, | |
| "learning_rate": 2.0942066466944355e-05, | |
| "loss": 0.5764, | |
| "mean_token_accuracy": 0.813313227891922, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 1.896755162241888, | |
| "grad_norm": 0.32508930563926697, | |
| "learning_rate": 1.9542666212969686e-05, | |
| "loss": 0.5767, | |
| "mean_token_accuracy": 0.8132544249296189, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.9557522123893807, | |
| "grad_norm": 0.26032930612564087, | |
| "learning_rate": 1.817762612003957e-05, | |
| "loss": 0.5782, | |
| "mean_token_accuracy": 0.8123401254415512, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 2.023598820058997, | |
| "grad_norm": 0.34573885798454285, | |
| "learning_rate": 1.685284049882144e-05, | |
| "loss": 0.6753, | |
| "mean_token_accuracy": 0.8178334352446766, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.0825958702064895, | |
| "grad_norm": 0.32736095786094666, | |
| "learning_rate": 1.5574029839182002e-05, | |
| "loss": 0.5317, | |
| "mean_token_accuracy": 0.8253544390201568, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.1415929203539825, | |
| "grad_norm": 0.27296292781829834, | |
| "learning_rate": 1.4346716108828856e-05, | |
| "loss": 0.5281, | |
| "mean_token_accuracy": 0.8263761848211288, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.200589970501475, | |
| "grad_norm": 0.29768332839012146, | |
| "learning_rate": 1.3176198909180748e-05, | |
| "loss": 0.53, | |
| "mean_token_accuracy": 0.8254351049661637, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 2.2595870206489677, | |
| "grad_norm": 0.2496112883090973, | |
| "learning_rate": 1.2067532591426361e-05, | |
| "loss": 0.5294, | |
| "mean_token_accuracy": 0.8257370889186859, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.3185840707964602, | |
| "grad_norm": 0.23834161460399628, | |
| "learning_rate": 1.1025504431585829e-05, | |
| "loss": 0.5343, | |
| "mean_token_accuracy": 0.8240474551916123, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 2.377581120943953, | |
| "grad_norm": 0.2458484023809433, | |
| "learning_rate": 1.0054613958815923e-05, | |
| "loss": 0.5311, | |
| "mean_token_accuracy": 0.8251578345894813, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.4365781710914454, | |
| "grad_norm": 0.22741276025772095, | |
| "learning_rate": 9.159053526220584e-06, | |
| "loss": 0.5274, | |
| "mean_token_accuracy": 0.826587364077568, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 2.495575221238938, | |
| "grad_norm": 0.23274612426757812, | |
| "learning_rate": 8.342690208062693e-06, | |
| "loss": 0.5356, | |
| "mean_token_accuracy": 0.8239503756165505, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.554572271386431, | |
| "grad_norm": 0.24258707463741302, | |
| "learning_rate": 7.609049101546113e-06, | |
| "loss": 0.5346, | |
| "mean_token_accuracy": 0.8239579856395721, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 2.613569321533923, | |
| "grad_norm": 0.2307581752538681, | |
| "learning_rate": 6.961298105271478e-06, | |
| "loss": 0.5309, | |
| "mean_token_accuracy": 0.8249035343527794, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.672566371681416, | |
| "grad_norm": 0.22489875555038452, | |
| "learning_rate": 6.40223424009332e-06, | |
| "loss": 0.5329, | |
| "mean_token_accuracy": 0.8248426854610443, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 2.7315634218289087, | |
| "grad_norm": 0.2095792293548584, | |
| "learning_rate": 5.934271571445601e-06, | |
| "loss": 0.5286, | |
| "mean_token_accuracy": 0.8259800374507904, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.7905604719764012, | |
| "grad_norm": 0.2478594034910202, | |
| "learning_rate": 5.559430785287692e-06, | |
| "loss": 0.5322, | |
| "mean_token_accuracy": 0.8249810084700584, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 2.849557522123894, | |
| "grad_norm": 0.21585014462471008, | |
| "learning_rate": 5.279330462682451e-06, | |
| "loss": 0.5315, | |
| "mean_token_accuracy": 0.8252547547221184, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.9085545722713864, | |
| "grad_norm": 0.21936820447444916, | |
| "learning_rate": 5.095180090683062e-06, | |
| "loss": 0.5322, | |
| "mean_token_accuracy": 0.8245489120483398, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 2.967551622418879, | |
| "grad_norm": 0.20714396238327026, | |
| "learning_rate": 5.007774839708124e-06, | |
| "loss": 0.5181, | |
| "mean_token_accuracy": 0.8292045801877975, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.991150442477876, | |
| "step": 252, | |
| "total_flos": 383335494844416.0, | |
| "train_loss": 0.0, | |
| "train_runtime": 1.7606, | |
| "train_samples_per_second": 18380.411, | |
| "train_steps_per_second": 143.131 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 252, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 383335494844416.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |