| { | |
| "best_metric": 0.8361669242658424, | |
| "best_model_checkpoint": "mit-b2-fv-finetuned-memes/checkpoint-180", | |
| "epoch": 19.987654320987655, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 3e-05, | |
| "loss": 1.6187, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 6e-05, | |
| "loss": 1.3683, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.5703245749613601, | |
| "eval_f1": 0.4915430460832317, | |
| "eval_loss": 1.1798477172851562, | |
| "eval_precision": 0.4913898246935893, | |
| "eval_recall": 0.5703245749613601, | |
| "eval_runtime": 25.7297, | |
| "eval_samples_per_second": 50.292, | |
| "eval_steps_per_second": 0.816, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 9e-05, | |
| "loss": 1.1618, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.00012, | |
| "loss": 1.0113, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_accuracy": 0.615919629057187, | |
| "eval_f1": 0.6273659431894075, | |
| "eval_loss": 1.0383963584899902, | |
| "eval_precision": 0.6812645697652936, | |
| "eval_recall": 0.615919629057187, | |
| "eval_runtime": 24.3501, | |
| "eval_samples_per_second": 53.142, | |
| "eval_steps_per_second": 0.862, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 0.00011666666666666667, | |
| "loss": 0.89, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 0.00011333333333333333, | |
| "loss": 0.7581, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.6808346213292118, | |
| "eval_f1": 0.6839700359928248, | |
| "eval_loss": 0.834769070148468, | |
| "eval_precision": 0.7376556278142583, | |
| "eval_recall": 0.6808346213292118, | |
| "eval_runtime": 25.9343, | |
| "eval_samples_per_second": 49.895, | |
| "eval_steps_per_second": 0.81, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 0.00011, | |
| "loss": 0.706, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 0.00010666666666666667, | |
| "loss": 0.6241, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "eval_accuracy": 0.7712519319938176, | |
| "eval_f1": 0.7734801228569064, | |
| "eval_loss": 0.6034244894981384, | |
| "eval_precision": 0.7864230813661904, | |
| "eval_recall": 0.7712519319938176, | |
| "eval_runtime": 24.0546, | |
| "eval_samples_per_second": 53.794, | |
| "eval_steps_per_second": 0.873, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 0.00010333333333333334, | |
| "loss": 0.574, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4999, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "eval_accuracy": 0.794435857805255, | |
| "eval_f1": 0.7908928134527277, | |
| "eval_loss": 0.5480836033821106, | |
| "eval_precision": 0.7999661256925431, | |
| "eval_recall": 0.794435857805255, | |
| "eval_runtime": 24.7472, | |
| "eval_samples_per_second": 52.289, | |
| "eval_steps_per_second": 0.849, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "learning_rate": 9.666666666666667e-05, | |
| "loss": 0.4429, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "learning_rate": 9.333333333333334e-05, | |
| "loss": 0.3981, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "eval_accuracy": 0.8021638330757341, | |
| "eval_f1": 0.8000125861001945, | |
| "eval_loss": 0.5253472328186035, | |
| "eval_precision": 0.8090800980369639, | |
| "eval_recall": 0.8021638330757341, | |
| "eval_runtime": 24.5494, | |
| "eval_samples_per_second": 52.71, | |
| "eval_steps_per_second": 0.855, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "learning_rate": 9e-05, | |
| "loss": 0.3603, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "learning_rate": 8.666666666666667e-05, | |
| "loss": 0.3484, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "eval_accuracy": 0.8238021638330757, | |
| "eval_f1": 0.8146387182540739, | |
| "eval_loss": 0.46875712275505066, | |
| "eval_precision": 0.8147156146167328, | |
| "eval_recall": 0.8238021638330757, | |
| "eval_runtime": 25.2779, | |
| "eval_samples_per_second": 51.191, | |
| "eval_steps_per_second": 0.831, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 7.49, | |
| "learning_rate": 8.333333333333333e-05, | |
| "loss": 0.2789, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 7.99, | |
| "learning_rate": 7.999999999999999e-05, | |
| "loss": 0.3142, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 7.99, | |
| "eval_accuracy": 0.7867078825347759, | |
| "eval_f1": 0.7919733028920879, | |
| "eval_loss": 0.6245487928390503, | |
| "eval_precision": 0.820948058010093, | |
| "eval_recall": 0.7867078825347759, | |
| "eval_runtime": 24.2688, | |
| "eval_samples_per_second": 53.32, | |
| "eval_steps_per_second": 0.865, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "learning_rate": 7.666666666666667e-05, | |
| "loss": 0.2513, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "learning_rate": 7.333333333333334e-05, | |
| "loss": 0.2339, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "eval_accuracy": 0.8361669242658424, | |
| "eval_f1": 0.8354791396567843, | |
| "eval_loss": 0.5053289532661438, | |
| "eval_precision": 0.8426050546923035, | |
| "eval_recall": 0.8361669242658424, | |
| "eval_runtime": 24.696, | |
| "eval_samples_per_second": 52.397, | |
| "eval_steps_per_second": 0.85, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 9.49, | |
| "learning_rate": 7.000000000000001e-05, | |
| "loss": 0.1999, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 9.99, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.2284, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.99, | |
| "eval_accuracy": 0.8230293663060279, | |
| "eval_f1": 0.8187153015149123, | |
| "eval_loss": 0.5069507360458374, | |
| "eval_precision": 0.822032270944375, | |
| "eval_recall": 0.8230293663060279, | |
| "eval_runtime": 23.6165, | |
| "eval_samples_per_second": 54.792, | |
| "eval_steps_per_second": 0.889, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 10.49, | |
| "learning_rate": 6.333333333333333e-05, | |
| "loss": 0.1861, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 10.99, | |
| "learning_rate": 6e-05, | |
| "loss": 0.1824, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 10.99, | |
| "eval_accuracy": 0.8006182380216383, | |
| "eval_f1": 0.8035059555919015, | |
| "eval_loss": 0.5779785513877869, | |
| "eval_precision": 0.8138172496848511, | |
| "eval_recall": 0.8006182380216383, | |
| "eval_runtime": 24.9222, | |
| "eval_samples_per_second": 51.922, | |
| "eval_steps_per_second": 0.843, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 11.49, | |
| "learning_rate": 5.6666666666666664e-05, | |
| "loss": 0.1647, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 11.99, | |
| "learning_rate": 5.333333333333333e-05, | |
| "loss": 0.1561, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 11.99, | |
| "eval_accuracy": 0.8253477588871716, | |
| "eval_f1": 0.8217716611197545, | |
| "eval_loss": 0.5429410338401794, | |
| "eval_precision": 0.8196794558105368, | |
| "eval_recall": 0.8253477588871716, | |
| "eval_runtime": 23.8337, | |
| "eval_samples_per_second": 54.293, | |
| "eval_steps_per_second": 0.881, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 12.49, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1551, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "learning_rate": 4.666666666666667e-05, | |
| "loss": 0.1229, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "eval_accuracy": 0.8330757341576507, | |
| "eval_f1": 0.8303358084478046, | |
| "eval_loss": 0.5324836373329163, | |
| "eval_precision": 0.8296069273511578, | |
| "eval_recall": 0.8330757341576507, | |
| "eval_runtime": 25.348, | |
| "eval_samples_per_second": 51.049, | |
| "eval_steps_per_second": 0.828, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 13.49, | |
| "learning_rate": 4.3333333333333334e-05, | |
| "loss": 0.1208, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 13.99, | |
| "learning_rate": 3.9999999999999996e-05, | |
| "loss": 0.1232, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 13.99, | |
| "eval_accuracy": 0.8276661514683153, | |
| "eval_f1": 0.8273115902224707, | |
| "eval_loss": 0.5595067143440247, | |
| "eval_precision": 0.8290015047050906, | |
| "eval_recall": 0.8276661514683153, | |
| "eval_runtime": 23.6014, | |
| "eval_samples_per_second": 54.827, | |
| "eval_steps_per_second": 0.89, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 14.49, | |
| "learning_rate": 3.666666666666667e-05, | |
| "loss": 0.1204, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 14.99, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.118, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 14.99, | |
| "eval_accuracy": 0.8292117465224111, | |
| "eval_f1": 0.8298744039909668, | |
| "eval_loss": 0.5973792672157288, | |
| "eval_precision": 0.8344810455815268, | |
| "eval_recall": 0.8292117465224111, | |
| "eval_runtime": 24.4918, | |
| "eval_samples_per_second": 52.834, | |
| "eval_steps_per_second": 0.857, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 15.49, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1015, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 15.99, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 0.11, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 15.99, | |
| "eval_accuracy": 0.8253477588871716, | |
| "eval_f1": 0.8230916961516846, | |
| "eval_loss": 0.579598069190979, | |
| "eval_precision": 0.8228234989922505, | |
| "eval_recall": 0.8253477588871716, | |
| "eval_runtime": 23.8333, | |
| "eval_samples_per_second": 54.294, | |
| "eval_steps_per_second": 0.881, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 16.49, | |
| "learning_rate": 2.3333333333333336e-05, | |
| "loss": 0.1037, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "learning_rate": 1.9999999999999998e-05, | |
| "loss": 0.0948, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "eval_accuracy": 0.8346213292117465, | |
| "eval_f1": 0.8348916431445179, | |
| "eval_loss": 0.5581147074699402, | |
| "eval_precision": 0.8357545769977985, | |
| "eval_recall": 0.8346213292117465, | |
| "eval_runtime": 24.5732, | |
| "eval_samples_per_second": 52.659, | |
| "eval_steps_per_second": 0.855, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 17.49, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0933, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 17.99, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.0985, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 17.99, | |
| "eval_accuracy": 0.8338485316846986, | |
| "eval_f1": 0.8318239397011512, | |
| "eval_loss": 0.569961428642273, | |
| "eval_precision": 0.830062297595451, | |
| "eval_recall": 0.8338485316846986, | |
| "eval_runtime": 24.4249, | |
| "eval_samples_per_second": 52.979, | |
| "eval_steps_per_second": 0.86, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 18.49, | |
| "learning_rate": 9.999999999999999e-06, | |
| "loss": 0.0877, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.0821, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "eval_accuracy": 0.8330757341576507, | |
| "eval_f1": 0.833525849625881, | |
| "eval_loss": 0.5755681395530701, | |
| "eval_precision": 0.8342801097840022, | |
| "eval_recall": 0.8330757341576507, | |
| "eval_runtime": 23.5787, | |
| "eval_samples_per_second": 54.88, | |
| "eval_steps_per_second": 0.891, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 19.49, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.0843, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 19.99, | |
| "learning_rate": 0.0, | |
| "loss": 0.0813, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.99, | |
| "eval_accuracy": 0.8323029366306027, | |
| "eval_f1": 0.831492653119617, | |
| "eval_loss": 0.5984169840812683, | |
| "eval_precision": 0.831217385971583, | |
| "eval_recall": 0.8323029366306027, | |
| "eval_runtime": 24.9692, | |
| "eval_samples_per_second": 51.824, | |
| "eval_steps_per_second": 0.841, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.99, | |
| "step": 400, | |
| "total_flos": 1.1809647563061068e+19, | |
| "train_loss": 0.36638923436403276, | |
| "train_runtime": 5624.3012, | |
| "train_samples_per_second": 18.399, | |
| "train_steps_per_second": 0.071 | |
| } | |
| ], | |
| "max_steps": 400, | |
| "num_train_epochs": 20, | |
| "total_flos": 1.1809647563061068e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |