| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.23143277929728592, | |
| "eval_steps": 10, | |
| "global_step": 550, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004207868714496108, | |
| "grad_norm": 3.2926599979400635, | |
| "learning_rate": 9.818181818181818e-05, | |
| "loss": 0.6744, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.004207868714496108, | |
| "eval_accuracy": 0.6254996657371521, | |
| "eval_loss": 0.6731761693954468, | |
| "eval_runtime": 574.0292, | |
| "eval_samples_per_second": 8.28, | |
| "eval_steps_per_second": 2.071, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008415737428992216, | |
| "grad_norm": 3.0475454330444336, | |
| "learning_rate": 9.636363636363637e-05, | |
| "loss": 0.6978, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.008415737428992216, | |
| "eval_accuracy": 0.6509572863578796, | |
| "eval_loss": 0.6372496485710144, | |
| "eval_runtime": 563.1409, | |
| "eval_samples_per_second": 8.44, | |
| "eval_steps_per_second": 2.111, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.012623606143488323, | |
| "grad_norm": 13.133508682250977, | |
| "learning_rate": 9.454545454545455e-05, | |
| "loss": 0.6593, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.012623606143488323, | |
| "eval_accuracy": 0.8199031949043274, | |
| "eval_loss": 0.5605600476264954, | |
| "eval_runtime": 570.6751, | |
| "eval_samples_per_second": 8.329, | |
| "eval_steps_per_second": 2.083, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.016831474857984433, | |
| "grad_norm": 19.010900497436523, | |
| "learning_rate": 9.272727272727273e-05, | |
| "loss": 0.5335, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.016831474857984433, | |
| "eval_accuracy": 0.7992846369743347, | |
| "eval_loss": 0.5063626766204834, | |
| "eval_runtime": 573.7613, | |
| "eval_samples_per_second": 8.284, | |
| "eval_steps_per_second": 2.072, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.021039343572480537, | |
| "grad_norm": 8.88330078125, | |
| "learning_rate": 9.090909090909092e-05, | |
| "loss": 0.5275, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.021039343572480537, | |
| "eval_accuracy": 0.8531453609466553, | |
| "eval_loss": 0.42206627130508423, | |
| "eval_runtime": 570.9439, | |
| "eval_samples_per_second": 8.325, | |
| "eval_steps_per_second": 2.083, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.025247212286976645, | |
| "grad_norm": 7.115237236022949, | |
| "learning_rate": 8.90909090909091e-05, | |
| "loss": 0.332, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.025247212286976645, | |
| "eval_accuracy": 0.8525142073631287, | |
| "eval_loss": 0.3472881615161896, | |
| "eval_runtime": 570.3915, | |
| "eval_samples_per_second": 8.333, | |
| "eval_steps_per_second": 2.085, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.029455081001472753, | |
| "grad_norm": 4.491124629974365, | |
| "learning_rate": 8.727272727272727e-05, | |
| "loss": 0.3109, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.029455081001472753, | |
| "eval_accuracy": 0.8998527526855469, | |
| "eval_loss": 0.24124681949615479, | |
| "eval_runtime": 571.9368, | |
| "eval_samples_per_second": 8.31, | |
| "eval_steps_per_second": 2.079, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.033662949715968865, | |
| "grad_norm": 5.824629306793213, | |
| "learning_rate": 8.545454545454545e-05, | |
| "loss": 0.3167, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.033662949715968865, | |
| "eval_accuracy": 0.9421418309211731, | |
| "eval_loss": 0.16658170521259308, | |
| "eval_runtime": 581.5044, | |
| "eval_samples_per_second": 8.174, | |
| "eval_steps_per_second": 2.045, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.03787081843046497, | |
| "grad_norm": 14.90581226348877, | |
| "learning_rate": 8.363636363636364e-05, | |
| "loss": 0.441, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.03787081843046497, | |
| "eval_accuracy": 0.9120555520057678, | |
| "eval_loss": 0.24102848768234253, | |
| "eval_runtime": 575.9749, | |
| "eval_samples_per_second": 8.252, | |
| "eval_steps_per_second": 2.064, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.042078687144961074, | |
| "grad_norm": 5.160868167877197, | |
| "learning_rate": 8.181818181818183e-05, | |
| "loss": 0.3689, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.042078687144961074, | |
| "eval_accuracy": 0.9105827808380127, | |
| "eval_loss": 0.24303115904331207, | |
| "eval_runtime": 566.896, | |
| "eval_samples_per_second": 8.384, | |
| "eval_steps_per_second": 2.097, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.046286555859457186, | |
| "grad_norm": 4.731154918670654, | |
| "learning_rate": 8e-05, | |
| "loss": 0.2679, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.046286555859457186, | |
| "eval_accuracy": 0.9086892604827881, | |
| "eval_loss": 0.261859655380249, | |
| "eval_runtime": 566.3825, | |
| "eval_samples_per_second": 8.392, | |
| "eval_steps_per_second": 2.099, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.05049442457395329, | |
| "grad_norm": 3.6892170906066895, | |
| "learning_rate": 7.818181818181818e-05, | |
| "loss": 0.2681, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.05049442457395329, | |
| "eval_accuracy": 0.8956448435783386, | |
| "eval_loss": 0.31176039576530457, | |
| "eval_runtime": 571.9255, | |
| "eval_samples_per_second": 8.311, | |
| "eval_steps_per_second": 2.079, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.0547022932884494, | |
| "grad_norm": 0.8576626181602478, | |
| "learning_rate": 7.636363636363637e-05, | |
| "loss": 0.3362, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.0547022932884494, | |
| "eval_accuracy": 0.9206816554069519, | |
| "eval_loss": 0.23810358345508575, | |
| "eval_runtime": 565.3779, | |
| "eval_samples_per_second": 8.407, | |
| "eval_steps_per_second": 2.103, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.05891016200294551, | |
| "grad_norm": 0.5792973637580872, | |
| "learning_rate": 7.454545454545455e-05, | |
| "loss": 0.2112, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.05891016200294551, | |
| "eval_accuracy": 0.9688617587089539, | |
| "eval_loss": 0.09994017332792282, | |
| "eval_runtime": 565.2596, | |
| "eval_samples_per_second": 8.409, | |
| "eval_steps_per_second": 2.103, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.06311803071744161, | |
| "grad_norm": 13.7179594039917, | |
| "learning_rate": 7.272727272727273e-05, | |
| "loss": 0.2842, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.06311803071744161, | |
| "eval_accuracy": 0.9562381505966187, | |
| "eval_loss": 0.13098464906215668, | |
| "eval_runtime": 568.883, | |
| "eval_samples_per_second": 8.355, | |
| "eval_steps_per_second": 2.09, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.06732589943193773, | |
| "grad_norm": 13.58166217803955, | |
| "learning_rate": 7.090909090909092e-05, | |
| "loss": 0.2738, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.06732589943193773, | |
| "eval_accuracy": 0.974121630191803, | |
| "eval_loss": 0.07404066622257233, | |
| "eval_runtime": 565.2694, | |
| "eval_samples_per_second": 8.408, | |
| "eval_steps_per_second": 2.103, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.07153376814643383, | |
| "grad_norm": 1.1847424507141113, | |
| "learning_rate": 6.90909090909091e-05, | |
| "loss": 0.3975, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.07153376814643383, | |
| "eval_accuracy": 0.9358299970626831, | |
| "eval_loss": 0.1807854324579239, | |
| "eval_runtime": 565.7049, | |
| "eval_samples_per_second": 8.402, | |
| "eval_steps_per_second": 2.102, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.07574163686092994, | |
| "grad_norm": 0.5569546818733215, | |
| "learning_rate": 6.727272727272727e-05, | |
| "loss": 0.2441, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.07574163686092994, | |
| "eval_accuracy": 0.9688617587089539, | |
| "eval_loss": 0.09184221178293228, | |
| "eval_runtime": 555.8532, | |
| "eval_samples_per_second": 8.551, | |
| "eval_steps_per_second": 2.139, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.07994950557542604, | |
| "grad_norm": 5.223395347595215, | |
| "learning_rate": 6.545454545454546e-05, | |
| "loss": 0.15, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.07994950557542604, | |
| "eval_accuracy": 0.9474016427993774, | |
| "eval_loss": 0.22705134749412537, | |
| "eval_runtime": 554.1797, | |
| "eval_samples_per_second": 8.577, | |
| "eval_steps_per_second": 2.146, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.08415737428992215, | |
| "grad_norm": 19.633956909179688, | |
| "learning_rate": 6.363636363636364e-05, | |
| "loss": 0.4032, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08415737428992215, | |
| "eval_accuracy": 0.9152114391326904, | |
| "eval_loss": 0.3923904001712799, | |
| "eval_runtime": 554.483, | |
| "eval_samples_per_second": 8.572, | |
| "eval_steps_per_second": 2.144, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08836524300441827, | |
| "grad_norm": 0.19549815356731415, | |
| "learning_rate": 6.181818181818182e-05, | |
| "loss": 0.0958, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.08836524300441827, | |
| "eval_accuracy": 0.9160529971122742, | |
| "eval_loss": 0.3585941791534424, | |
| "eval_runtime": 546.3632, | |
| "eval_samples_per_second": 8.699, | |
| "eval_steps_per_second": 2.176, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.09257311171891437, | |
| "grad_norm": 1.1258805990219116, | |
| "learning_rate": 6e-05, | |
| "loss": 0.2203, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.09257311171891437, | |
| "eval_accuracy": 0.895013689994812, | |
| "eval_loss": 0.3808138072490692, | |
| "eval_runtime": 549.8489, | |
| "eval_samples_per_second": 8.644, | |
| "eval_steps_per_second": 2.162, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.09678098043341048, | |
| "grad_norm": 2.5008387565612793, | |
| "learning_rate": 5.818181818181818e-05, | |
| "loss": 0.0868, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.09678098043341048, | |
| "eval_accuracy": 0.9440353512763977, | |
| "eval_loss": 0.21156248450279236, | |
| "eval_runtime": 554.8432, | |
| "eval_samples_per_second": 8.566, | |
| "eval_steps_per_second": 2.143, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.10098884914790658, | |
| "grad_norm": 0.5708354115486145, | |
| "learning_rate": 5.636363636363636e-05, | |
| "loss": 0.1104, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.10098884914790658, | |
| "eval_accuracy": 0.9168946146965027, | |
| "eval_loss": 0.3665996193885803, | |
| "eval_runtime": 557.2932, | |
| "eval_samples_per_second": 8.529, | |
| "eval_steps_per_second": 2.134, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.1051967178624027, | |
| "grad_norm": 0.9385812878608704, | |
| "learning_rate": 5.4545454545454546e-05, | |
| "loss": 0.211, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.1051967178624027, | |
| "eval_accuracy": 0.9593940377235413, | |
| "eval_loss": 0.1287224441766739, | |
| "eval_runtime": 553.8519, | |
| "eval_samples_per_second": 8.582, | |
| "eval_steps_per_second": 2.147, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.1094045865768988, | |
| "grad_norm": 5.300781726837158, | |
| "learning_rate": 5.272727272727272e-05, | |
| "loss": 0.3359, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.1094045865768988, | |
| "eval_accuracy": 0.9863244295120239, | |
| "eval_loss": 0.04394599795341492, | |
| "eval_runtime": 553.7785, | |
| "eval_samples_per_second": 8.583, | |
| "eval_steps_per_second": 2.147, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.11361245529139491, | |
| "grad_norm": 6.374191761016846, | |
| "learning_rate": 5.090909090909091e-05, | |
| "loss": 0.2681, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.11361245529139491, | |
| "eval_accuracy": 0.9440353512763977, | |
| "eval_loss": 0.16353961825370789, | |
| "eval_runtime": 553.8134, | |
| "eval_samples_per_second": 8.582, | |
| "eval_steps_per_second": 2.147, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.11782032400589101, | |
| "grad_norm": 0.06790979206562042, | |
| "learning_rate": 4.909090909090909e-05, | |
| "loss": 0.0586, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.11782032400589101, | |
| "eval_accuracy": 0.9688617587089539, | |
| "eval_loss": 0.08115343004465103, | |
| "eval_runtime": 552.432, | |
| "eval_samples_per_second": 8.604, | |
| "eval_steps_per_second": 2.152, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.12202819272038712, | |
| "grad_norm": 2.017587423324585, | |
| "learning_rate": 4.7272727272727275e-05, | |
| "loss": 0.215, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.12202819272038712, | |
| "eval_accuracy": 0.9255207180976868, | |
| "eval_loss": 0.22139500081539154, | |
| "eval_runtime": 552.6796, | |
| "eval_samples_per_second": 8.6, | |
| "eval_steps_per_second": 2.151, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.12623606143488322, | |
| "grad_norm": 0.08541750907897949, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 0.1846, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.12623606143488322, | |
| "eval_accuracy": 0.9837996959686279, | |
| "eval_loss": 0.049501266330480576, | |
| "eval_runtime": 551.6552, | |
| "eval_samples_per_second": 8.616, | |
| "eval_steps_per_second": 2.155, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.13044393014937933, | |
| "grad_norm": 14.675045013427734, | |
| "learning_rate": 4.3636363636363636e-05, | |
| "loss": 0.2153, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.13044393014937933, | |
| "eval_accuracy": 0.9835892915725708, | |
| "eval_loss": 0.05814049392938614, | |
| "eval_runtime": 553.8442, | |
| "eval_samples_per_second": 8.582, | |
| "eval_steps_per_second": 2.147, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.13465179886387546, | |
| "grad_norm": 0.5145708918571472, | |
| "learning_rate": 4.181818181818182e-05, | |
| "loss": 0.0814, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.13465179886387546, | |
| "eval_accuracy": 0.9575005173683167, | |
| "eval_loss": 0.1593790054321289, | |
| "eval_runtime": 554.2522, | |
| "eval_samples_per_second": 8.576, | |
| "eval_steps_per_second": 2.145, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.13885966757837157, | |
| "grad_norm": 15.138551712036133, | |
| "learning_rate": 4e-05, | |
| "loss": 0.1322, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.13885966757837157, | |
| "eval_accuracy": 0.9673890471458435, | |
| "eval_loss": 0.11336074024438858, | |
| "eval_runtime": 548.0702, | |
| "eval_samples_per_second": 8.672, | |
| "eval_steps_per_second": 2.169, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.14306753629286767, | |
| "grad_norm": 0.017534606158733368, | |
| "learning_rate": 3.818181818181819e-05, | |
| "loss": 0.1172, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.14306753629286767, | |
| "eval_accuracy": 0.9604460597038269, | |
| "eval_loss": 0.11512165516614914, | |
| "eval_runtime": 551.6579, | |
| "eval_samples_per_second": 8.616, | |
| "eval_steps_per_second": 2.155, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.14727540500736377, | |
| "grad_norm": 0.008281617425382137, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 0.0648, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.14727540500736377, | |
| "eval_accuracy": 0.9758047461509705, | |
| "eval_loss": 0.06669025123119354, | |
| "eval_runtime": 553.29, | |
| "eval_samples_per_second": 8.59, | |
| "eval_steps_per_second": 2.149, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.15148327372185988, | |
| "grad_norm": 0.17740336060523987, | |
| "learning_rate": 3.454545454545455e-05, | |
| "loss": 0.2261, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.15148327372185988, | |
| "eval_accuracy": 0.967809796333313, | |
| "eval_loss": 0.09138375520706177, | |
| "eval_runtime": 561.1917, | |
| "eval_samples_per_second": 8.469, | |
| "eval_steps_per_second": 2.119, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.15569114243635598, | |
| "grad_norm": 1.9274922609329224, | |
| "learning_rate": 3.272727272727273e-05, | |
| "loss": 0.1405, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.15569114243635598, | |
| "eval_accuracy": 0.9215232729911804, | |
| "eval_loss": 0.294222891330719, | |
| "eval_runtime": 554.2437, | |
| "eval_samples_per_second": 8.576, | |
| "eval_steps_per_second": 2.145, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.1598990111508521, | |
| "grad_norm": 20.162691116333008, | |
| "learning_rate": 3.090909090909091e-05, | |
| "loss": 0.2011, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.1598990111508521, | |
| "eval_accuracy": 0.9770671129226685, | |
| "eval_loss": 0.06251883506774902, | |
| "eval_runtime": 556.1293, | |
| "eval_samples_per_second": 8.547, | |
| "eval_steps_per_second": 2.138, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.1641068798653482, | |
| "grad_norm": 0.0279800184071064, | |
| "learning_rate": 2.909090909090909e-05, | |
| "loss": 0.1612, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.1641068798653482, | |
| "eval_accuracy": 0.9840101003646851, | |
| "eval_loss": 0.040968313813209534, | |
| "eval_runtime": 554.0799, | |
| "eval_samples_per_second": 8.578, | |
| "eval_steps_per_second": 2.146, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.1683147485798443, | |
| "grad_norm": 8.97774600982666, | |
| "learning_rate": 2.7272727272727273e-05, | |
| "loss": 0.1875, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.1683147485798443, | |
| "eval_accuracy": 0.974121630191803, | |
| "eval_loss": 0.06667405366897583, | |
| "eval_runtime": 553.8832, | |
| "eval_samples_per_second": 8.581, | |
| "eval_steps_per_second": 2.147, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.17252261729434043, | |
| "grad_norm": 0.09378823637962341, | |
| "learning_rate": 2.5454545454545454e-05, | |
| "loss": 0.0717, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.17252261729434043, | |
| "eval_accuracy": 0.9829581379890442, | |
| "eval_loss": 0.04144088178873062, | |
| "eval_runtime": 554.6341, | |
| "eval_samples_per_second": 8.57, | |
| "eval_steps_per_second": 2.144, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.17673048600883653, | |
| "grad_norm": 0.21639426052570343, | |
| "learning_rate": 2.3636363636363637e-05, | |
| "loss": 0.0524, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.17673048600883653, | |
| "eval_accuracy": 0.9886387586593628, | |
| "eval_loss": 0.0340532623231411, | |
| "eval_runtime": 554.2365, | |
| "eval_samples_per_second": 8.576, | |
| "eval_steps_per_second": 2.145, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.18093835472333264, | |
| "grad_norm": 0.07831648737192154, | |
| "learning_rate": 2.1818181818181818e-05, | |
| "loss": 0.0935, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.18093835472333264, | |
| "eval_accuracy": 0.9842205047607422, | |
| "eval_loss": 0.0411146804690361, | |
| "eval_runtime": 557.1919, | |
| "eval_samples_per_second": 8.53, | |
| "eval_steps_per_second": 2.134, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.18514622343782874, | |
| "grad_norm": 0.0247107595205307, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0521, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.18514622343782874, | |
| "eval_accuracy": 0.974121630191803, | |
| "eval_loss": 0.07654237747192383, | |
| "eval_runtime": 548.0314, | |
| "eval_samples_per_second": 8.673, | |
| "eval_steps_per_second": 2.17, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.18935409215232485, | |
| "grad_norm": 0.10471628606319427, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 0.0889, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.18935409215232485, | |
| "eval_accuracy": 0.9608668088912964, | |
| "eval_loss": 0.1275644451379776, | |
| "eval_runtime": 558.2261, | |
| "eval_samples_per_second": 8.514, | |
| "eval_steps_per_second": 2.13, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.19356196086682095, | |
| "grad_norm": 0.4881088435649872, | |
| "learning_rate": 1.6363636363636366e-05, | |
| "loss": 0.0387, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.19356196086682095, | |
| "eval_accuracy": 0.9667578339576721, | |
| "eval_loss": 0.098667673766613, | |
| "eval_runtime": 556.0323, | |
| "eval_samples_per_second": 8.548, | |
| "eval_steps_per_second": 2.138, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.19776982958131706, | |
| "grad_norm": 0.01689060404896736, | |
| "learning_rate": 1.4545454545454545e-05, | |
| "loss": 0.03, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.19776982958131706, | |
| "eval_accuracy": 0.9734904170036316, | |
| "eval_loss": 0.07727943360805511, | |
| "eval_runtime": 554.2432, | |
| "eval_samples_per_second": 8.576, | |
| "eval_steps_per_second": 2.145, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.20197769829581316, | |
| "grad_norm": 8.339999198913574, | |
| "learning_rate": 1.2727272727272727e-05, | |
| "loss": 0.0901, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.20197769829581316, | |
| "eval_accuracy": 0.9829581379890442, | |
| "eval_loss": 0.04694819450378418, | |
| "eval_runtime": 556.8586, | |
| "eval_samples_per_second": 8.535, | |
| "eval_steps_per_second": 2.135, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.20618556701030927, | |
| "grad_norm": 0.023077547550201416, | |
| "learning_rate": 1.0909090909090909e-05, | |
| "loss": 0.0873, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.20618556701030927, | |
| "eval_accuracy": 0.9808542132377625, | |
| "eval_loss": 0.0580894760787487, | |
| "eval_runtime": 554.8362, | |
| "eval_samples_per_second": 8.566, | |
| "eval_steps_per_second": 2.143, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.2103934357248054, | |
| "grad_norm": 0.014377394691109657, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 0.0201, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2103934357248054, | |
| "eval_accuracy": 0.9816957712173462, | |
| "eval_loss": 0.057598479092121124, | |
| "eval_runtime": 554.1331, | |
| "eval_samples_per_second": 8.577, | |
| "eval_steps_per_second": 2.146, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2146013044393015, | |
| "grad_norm": 0.12555623054504395, | |
| "learning_rate": 7.272727272727272e-06, | |
| "loss": 0.017, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.2146013044393015, | |
| "eval_accuracy": 0.9802230000495911, | |
| "eval_loss": 0.06528125703334808, | |
| "eval_runtime": 553.9243, | |
| "eval_samples_per_second": 8.581, | |
| "eval_steps_per_second": 2.147, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.2188091731537976, | |
| "grad_norm": 0.04903264343738556, | |
| "learning_rate": 5.4545454545454545e-06, | |
| "loss": 0.1464, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.2188091731537976, | |
| "eval_accuracy": 0.9749631881713867, | |
| "eval_loss": 0.07879046350717545, | |
| "eval_runtime": 556.0819, | |
| "eval_samples_per_second": 8.547, | |
| "eval_steps_per_second": 2.138, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.22301704186829371, | |
| "grad_norm": 0.016027677804231644, | |
| "learning_rate": 3.636363636363636e-06, | |
| "loss": 0.0105, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.22301704186829371, | |
| "eval_accuracy": 0.9758047461509705, | |
| "eval_loss": 0.07757497578859329, | |
| "eval_runtime": 557.9097, | |
| "eval_samples_per_second": 8.519, | |
| "eval_steps_per_second": 2.131, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.22722491058278982, | |
| "grad_norm": 0.08870552480220795, | |
| "learning_rate": 1.818181818181818e-06, | |
| "loss": 0.0124, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.22722491058278982, | |
| "eval_accuracy": 0.9774879217147827, | |
| "eval_loss": 0.07116981595754623, | |
| "eval_runtime": 561.0259, | |
| "eval_samples_per_second": 8.472, | |
| "eval_steps_per_second": 2.119, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.23143277929728592, | |
| "grad_norm": 7.751413822174072, | |
| "learning_rate": 0.0, | |
| "loss": 0.061, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.23143277929728592, | |
| "eval_accuracy": 0.9779086709022522, | |
| "eval_loss": 0.06926773488521576, | |
| "eval_runtime": 562.0676, | |
| "eval_samples_per_second": 8.456, | |
| "eval_steps_per_second": 2.115, | |
| "step": 550 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 550, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.208452205931052e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |