| { | |
| "best_metric": 0.8323029366306027, | |
| "best_model_checkpoint": "mit-b2-VF2-finetuned-memes/checkpoint-320", | |
| "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": 1.5055, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 6e-05, | |
| "loss": 1.3077, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.5548686244204019, | |
| "eval_f1": 0.528612348642737, | |
| "eval_loss": 1.1682769060134888, | |
| "eval_precision": 0.5621381932725592, | |
| "eval_recall": 0.5548686244204019, | |
| "eval_runtime": 24.8047, | |
| "eval_samples_per_second": 52.167, | |
| "eval_steps_per_second": 0.847, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 9e-05, | |
| "loss": 1.1418, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.00012, | |
| "loss": 0.9359, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_accuracy": 0.6731066460587326, | |
| "eval_f1": 0.6535186565920124, | |
| "eval_loss": 0.8572518825531006, | |
| "eval_precision": 0.6806723513626268, | |
| "eval_recall": 0.6731066460587326, | |
| "eval_runtime": 24.0008, | |
| "eval_samples_per_second": 53.915, | |
| "eval_steps_per_second": 0.875, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 0.00011666666666666667, | |
| "loss": 0.8433, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 0.00011333333333333333, | |
| "loss": 0.7219, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.7272024729520865, | |
| "eval_f1": 0.7245846570171044, | |
| "eval_loss": 0.7105928659439087, | |
| "eval_precision": 0.7358764523852896, | |
| "eval_recall": 0.7272024729520865, | |
| "eval_runtime": 24.3322, | |
| "eval_samples_per_second": 53.181, | |
| "eval_steps_per_second": 0.863, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 0.00011, | |
| "loss": 0.639, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 0.00010666666666666667, | |
| "loss": 0.6013, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "eval_accuracy": 0.7550231839258115, | |
| "eval_f1": 0.7557556279759768, | |
| "eval_loss": 0.6444658041000366, | |
| "eval_precision": 0.7685934069718168, | |
| "eval_recall": 0.7550231839258115, | |
| "eval_runtime": 23.9013, | |
| "eval_samples_per_second": 54.139, | |
| "eval_steps_per_second": 0.879, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 0.00010333333333333334, | |
| "loss": 0.4977, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "learning_rate": 0.0001, | |
| "loss": 0.5243, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "eval_accuracy": 0.7573415765069552, | |
| "eval_f1": 0.7583802674435394, | |
| "eval_loss": 0.6717090606689453, | |
| "eval_precision": 0.8077053699577921, | |
| "eval_recall": 0.7573415765069552, | |
| "eval_runtime": 24.2773, | |
| "eval_samples_per_second": 53.301, | |
| "eval_steps_per_second": 0.865, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "learning_rate": 9.666666666666667e-05, | |
| "loss": 0.4352, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "learning_rate": 9.333333333333334e-05, | |
| "loss": 0.4409, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "eval_accuracy": 0.8068006182380216, | |
| "eval_f1": 0.7989048233772276, | |
| "eval_loss": 0.531519889831543, | |
| "eval_precision": 0.8027399261022145, | |
| "eval_recall": 0.8068006182380216, | |
| "eval_runtime": 24.4159, | |
| "eval_samples_per_second": 52.998, | |
| "eval_steps_per_second": 0.86, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "learning_rate": 9e-05, | |
| "loss": 0.3684, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "learning_rate": 8.666666666666667e-05, | |
| "loss": 0.3325, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "eval_accuracy": 0.8230293663060279, | |
| "eval_f1": 0.815761738328179, | |
| "eval_loss": 0.5158553719520569, | |
| "eval_precision": 0.8236186656882729, | |
| "eval_recall": 0.8230293663060279, | |
| "eval_runtime": 24.8705, | |
| "eval_samples_per_second": 52.029, | |
| "eval_steps_per_second": 0.844, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 7.49, | |
| "learning_rate": 8.333333333333333e-05, | |
| "loss": 0.304, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 7.99, | |
| "learning_rate": 7.999999999999999e-05, | |
| "loss": 0.2719, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 7.99, | |
| "eval_accuracy": 0.821483771251932, | |
| "eval_f1": 0.8202048206608414, | |
| "eval_loss": 0.5249665379524231, | |
| "eval_precision": 0.8226765366850811, | |
| "eval_recall": 0.821483771251932, | |
| "eval_runtime": 24.5067, | |
| "eval_samples_per_second": 52.802, | |
| "eval_steps_per_second": 0.857, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "learning_rate": 7.666666666666667e-05, | |
| "loss": 0.2209, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "learning_rate": 7.333333333333334e-05, | |
| "loss": 0.242, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "eval_accuracy": 0.8276661514683153, | |
| "eval_f1": 0.8267595186206215, | |
| "eval_loss": 0.5087464451789856, | |
| "eval_precision": 0.8259545635423114, | |
| "eval_recall": 0.8276661514683153, | |
| "eval_runtime": 24.2235, | |
| "eval_samples_per_second": 53.419, | |
| "eval_steps_per_second": 0.867, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 9.49, | |
| "learning_rate": 7.000000000000001e-05, | |
| "loss": 0.2021, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 9.99, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.2247, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.99, | |
| "eval_accuracy": 0.821483771251932, | |
| "eval_f1": 0.8217910575245788, | |
| "eval_loss": 0.5312910079956055, | |
| "eval_precision": 0.8274625356055352, | |
| "eval_recall": 0.821483771251932, | |
| "eval_runtime": 23.8959, | |
| "eval_samples_per_second": 54.152, | |
| "eval_steps_per_second": 0.879, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 10.49, | |
| "learning_rate": 6.333333333333333e-05, | |
| "loss": 0.1864, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 10.99, | |
| "learning_rate": 6e-05, | |
| "loss": 0.1955, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 10.99, | |
| "eval_accuracy": 0.8129829984544049, | |
| "eval_f1": 0.8073030128905387, | |
| "eval_loss": 0.6167330741882324, | |
| "eval_precision": 0.8062473668401103, | |
| "eval_recall": 0.8129829984544049, | |
| "eval_runtime": 24.1623, | |
| "eval_samples_per_second": 53.555, | |
| "eval_steps_per_second": 0.869, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 11.49, | |
| "learning_rate": 5.6666666666666664e-05, | |
| "loss": 0.1668, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 11.99, | |
| "learning_rate": 5.333333333333333e-05, | |
| "loss": 0.1567, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 11.99, | |
| "eval_accuracy": 0.8168469860896446, | |
| "eval_f1": 0.8172720638114461, | |
| "eval_loss": 0.585921049118042, | |
| "eval_precision": 0.818523777219596, | |
| "eval_recall": 0.8168469860896446, | |
| "eval_runtime": 24.2627, | |
| "eval_samples_per_second": 53.333, | |
| "eval_steps_per_second": 0.866, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 12.49, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1458, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "learning_rate": 4.666666666666667e-05, | |
| "loss": 0.1479, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "eval_accuracy": 0.821483771251932, | |
| "eval_f1": 0.8178479998623709, | |
| "eval_loss": 0.5937514901161194, | |
| "eval_precision": 0.8169057022013864, | |
| "eval_recall": 0.821483771251932, | |
| "eval_runtime": 24.5791, | |
| "eval_samples_per_second": 52.646, | |
| "eval_steps_per_second": 0.854, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 13.49, | |
| "learning_rate": 4.3333333333333334e-05, | |
| "loss": 0.1242, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 13.99, | |
| "learning_rate": 3.9999999999999996e-05, | |
| "loss": 0.1241, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 13.99, | |
| "eval_accuracy": 0.8261205564142194, | |
| "eval_f1": 0.8238689741768054, | |
| "eval_loss": 0.6187295317649841, | |
| "eval_precision": 0.8233844424179447, | |
| "eval_recall": 0.8261205564142194, | |
| "eval_runtime": 23.6967, | |
| "eval_samples_per_second": 54.607, | |
| "eval_steps_per_second": 0.886, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 14.49, | |
| "learning_rate": 3.666666666666667e-05, | |
| "loss": 0.1133, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 14.99, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.1114, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 14.99, | |
| "eval_accuracy": 0.8261205564142194, | |
| "eval_f1": 0.8293208039295025, | |
| "eval_loss": 0.6419451832771301, | |
| "eval_precision": 0.8351235142682792, | |
| "eval_recall": 0.8261205564142194, | |
| "eval_runtime": 24.6341, | |
| "eval_samples_per_second": 52.529, | |
| "eval_steps_per_second": 0.852, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 15.49, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1134, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 15.99, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 0.1022, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 15.99, | |
| "eval_accuracy": 0.8323029366306027, | |
| "eval_f1": 0.8293880616510334, | |
| "eval_loss": 0.6321794986724854, | |
| "eval_precision": 0.8283975183256338, | |
| "eval_recall": 0.8323029366306027, | |
| "eval_runtime": 24.0004, | |
| "eval_samples_per_second": 53.916, | |
| "eval_steps_per_second": 0.875, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 16.49, | |
| "learning_rate": 2.3333333333333336e-05, | |
| "loss": 0.099, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "learning_rate": 1.9999999999999998e-05, | |
| "loss": 0.0941, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "eval_accuracy": 0.8268933539412674, | |
| "eval_f1": 0.8263211093818184, | |
| "eval_loss": 0.6594778895378113, | |
| "eval_precision": 0.826589853355648, | |
| "eval_recall": 0.8268933539412674, | |
| "eval_runtime": 24.9004, | |
| "eval_samples_per_second": 51.967, | |
| "eval_steps_per_second": 0.843, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 17.49, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0908, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 17.99, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.0935, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 17.99, | |
| "eval_accuracy": 0.8268933539412674, | |
| "eval_f1": 0.8236963320695928, | |
| "eval_loss": 0.6673519015312195, | |
| "eval_precision": 0.8217985118762411, | |
| "eval_recall": 0.8268933539412674, | |
| "eval_runtime": 24.9117, | |
| "eval_samples_per_second": 51.943, | |
| "eval_steps_per_second": 0.843, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 18.49, | |
| "learning_rate": 9.999999999999999e-06, | |
| "loss": 0.0893, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.089, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "eval_accuracy": 0.8253477588871716, | |
| "eval_f1": 0.8235241860756777, | |
| "eval_loss": 0.6533293724060059, | |
| "eval_precision": 0.8222313831803589, | |
| "eval_recall": 0.8253477588871716, | |
| "eval_runtime": 24.9409, | |
| "eval_samples_per_second": 51.883, | |
| "eval_steps_per_second": 0.842, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 19.49, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.0861, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 19.99, | |
| "learning_rate": 0.0, | |
| "loss": 0.0794, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.99, | |
| "eval_accuracy": 0.8307573415765069, | |
| "eval_f1": 0.8286939083150942, | |
| "eval_loss": 0.6546534895896912, | |
| "eval_precision": 0.8272186656187493, | |
| "eval_recall": 0.8307573415765069, | |
| "eval_runtime": 25.4487, | |
| "eval_samples_per_second": 50.847, | |
| "eval_steps_per_second": 0.825, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.99, | |
| "step": 400, | |
| "total_flos": 1.1809647563061068e+19, | |
| "train_loss": 0.35424828216433524, | |
| "train_runtime": 5673.0945, | |
| "train_samples_per_second": 18.24, | |
| "train_steps_per_second": 0.071 | |
| } | |
| ], | |
| "max_steps": 400, | |
| "num_train_epochs": 20, | |
| "total_flos": 1.1809647563061068e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |