| { | |
| "best_metric": 0.34081346423562414, | |
| "best_model_checkpoint": "videomae-base-finetuned-elderf1/checkpoint-73", | |
| "epoch": 9.0875, | |
| "eval_steps": 500, | |
| "global_step": 720, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 4.3531270027160645, | |
| "learning_rate": 0.0001388888888888889, | |
| "loss": 1.7775, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 5.537198543548584, | |
| "learning_rate": 0.0002777777777777778, | |
| "loss": 1.6877, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 5.151556968688965, | |
| "learning_rate": 0.0004166666666666667, | |
| "loss": 1.7997, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 5.214244365692139, | |
| "learning_rate": 0.0005555555555555556, | |
| "loss": 1.8051, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 1.6136474609375, | |
| "learning_rate": 0.0006944444444444445, | |
| "loss": 1.7065, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 5.423031330108643, | |
| "learning_rate": 0.0008333333333333334, | |
| "loss": 1.7922, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 3.5037877559661865, | |
| "learning_rate": 0.0009722222222222222, | |
| "loss": 1.7358, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_accuracy": 0.34081346423562414, | |
| "eval_loss": 1.692335844039917, | |
| "eval_runtime": 379.7283, | |
| "eval_samples_per_second": 3.755, | |
| "eval_steps_per_second": 0.237, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 3.8850014209747314, | |
| "learning_rate": 0.0009876543209876543, | |
| "loss": 1.7738, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 6.418526649475098, | |
| "learning_rate": 0.0009722222222222222, | |
| "loss": 1.8192, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 6.298271179199219, | |
| "learning_rate": 0.0009567901234567902, | |
| "loss": 1.7833, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 4.652337551116943, | |
| "learning_rate": 0.000941358024691358, | |
| "loss": 1.7534, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 3.369626522064209, | |
| "learning_rate": 0.000925925925925926, | |
| "loss": 1.7376, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 2.182107448577881, | |
| "learning_rate": 0.0009104938271604939, | |
| "loss": 1.6406, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "grad_norm": 3.763148069381714, | |
| "learning_rate": 0.0008950617283950618, | |
| "loss": 1.7163, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_accuracy": 0.3373071528751753, | |
| "eval_loss": 1.66623854637146, | |
| "eval_runtime": 421.0248, | |
| "eval_samples_per_second": 3.387, | |
| "eval_steps_per_second": 0.214, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "grad_norm": 4.109099864959717, | |
| "learning_rate": 0.0008796296296296296, | |
| "loss": 1.6734, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 3.7187507152557373, | |
| "learning_rate": 0.0008641975308641975, | |
| "loss": 1.7535, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 2.9596164226531982, | |
| "learning_rate": 0.0008487654320987654, | |
| "loss": 1.7282, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 5.126859188079834, | |
| "learning_rate": 0.0008333333333333334, | |
| "loss": 1.7021, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "grad_norm": 2.36877703666687, | |
| "learning_rate": 0.0008179012345679012, | |
| "loss": 1.6977, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 4.42868185043335, | |
| "learning_rate": 0.0008024691358024692, | |
| "loss": 1.6882, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "grad_norm": 3.4869942665100098, | |
| "learning_rate": 0.0007870370370370372, | |
| "loss": 1.7018, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "eval_accuracy": 0.34081346423562414, | |
| "eval_loss": 1.6378456354141235, | |
| "eval_runtime": 423.5487, | |
| "eval_samples_per_second": 3.367, | |
| "eval_steps_per_second": 0.212, | |
| "step": 219 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.7394754886627197, | |
| "learning_rate": 0.0007716049382716049, | |
| "loss": 1.7356, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "grad_norm": 1.3488916158676147, | |
| "learning_rate": 0.0007561728395061729, | |
| "loss": 1.7024, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "grad_norm": 3.196521282196045, | |
| "learning_rate": 0.0007407407407407407, | |
| "loss": 1.7094, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "grad_norm": 3.9984352588653564, | |
| "learning_rate": 0.0007253086419753087, | |
| "loss": 1.6481, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "grad_norm": 3.6886866092681885, | |
| "learning_rate": 0.0007098765432098766, | |
| "loss": 1.652, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "grad_norm": 2.245149850845337, | |
| "learning_rate": 0.0006944444444444445, | |
| "loss": 1.7479, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "grad_norm": 4.62326192855835, | |
| "learning_rate": 0.0006790123456790124, | |
| "loss": 1.7129, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 4.474867343902588, | |
| "learning_rate": 0.0006635802469135802, | |
| "loss": 1.7334, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "eval_accuracy": 0.34011220196353437, | |
| "eval_loss": 1.6562532186508179, | |
| "eval_runtime": 383.4418, | |
| "eval_samples_per_second": 3.719, | |
| "eval_steps_per_second": 0.235, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "grad_norm": 1.4795947074890137, | |
| "learning_rate": 0.0006481481481481481, | |
| "loss": 1.7427, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "grad_norm": 2.173116683959961, | |
| "learning_rate": 0.0006327160493827161, | |
| "loss": 1.6894, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "grad_norm": 2.731816291809082, | |
| "learning_rate": 0.0006172839506172839, | |
| "loss": 1.6818, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "grad_norm": 2.120103120803833, | |
| "learning_rate": 0.0006018518518518519, | |
| "loss": 1.6782, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "grad_norm": 3.7916502952575684, | |
| "learning_rate": 0.0005864197530864199, | |
| "loss": 1.7388, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "grad_norm": 2.852003812789917, | |
| "learning_rate": 0.0005709876543209876, | |
| "loss": 1.7422, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "grad_norm": 6.300606727600098, | |
| "learning_rate": 0.0005555555555555556, | |
| "loss": 1.672, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "eval_accuracy": 0.23983169705469845, | |
| "eval_loss": 1.6567574739456177, | |
| "eval_runtime": 310.9306, | |
| "eval_samples_per_second": 4.586, | |
| "eval_steps_per_second": 0.289, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "grad_norm": 5.469892501831055, | |
| "learning_rate": 0.0005401234567901234, | |
| "loss": 1.6958, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "grad_norm": 3.4026269912719727, | |
| "learning_rate": 0.0005246913580246914, | |
| "loss": 1.7383, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "grad_norm": 2.7335870265960693, | |
| "learning_rate": 0.0005092592592592593, | |
| "loss": 1.6763, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "grad_norm": 4.079995155334473, | |
| "learning_rate": 0.0004938271604938272, | |
| "loss": 1.693, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "grad_norm": 2.8691930770874023, | |
| "learning_rate": 0.0004783950617283951, | |
| "loss": 1.7293, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "grad_norm": 2.4010772705078125, | |
| "learning_rate": 0.000462962962962963, | |
| "loss": 1.6812, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "grad_norm": 4.233634948730469, | |
| "learning_rate": 0.0004475308641975309, | |
| "loss": 1.7095, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "eval_accuracy": 0.3387096774193548, | |
| "eval_loss": 1.6313395500183105, | |
| "eval_runtime": 306.7441, | |
| "eval_samples_per_second": 4.649, | |
| "eval_steps_per_second": 0.293, | |
| "step": 438 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.844386100769043, | |
| "learning_rate": 0.00043209876543209873, | |
| "loss": 1.6502, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "grad_norm": 3.7761685848236084, | |
| "learning_rate": 0.0004166666666666667, | |
| "loss": 1.6892, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "grad_norm": 2.8903658390045166, | |
| "learning_rate": 0.0004012345679012346, | |
| "loss": 1.6006, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "grad_norm": 2.548739194869995, | |
| "learning_rate": 0.00038580246913580245, | |
| "loss": 1.7113, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "grad_norm": 3.1980948448181152, | |
| "learning_rate": 0.00037037037037037035, | |
| "loss": 1.7259, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "grad_norm": 3.119049549102783, | |
| "learning_rate": 0.0003549382716049383, | |
| "loss": 1.6634, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "grad_norm": 3.465067148208618, | |
| "learning_rate": 0.0003395061728395062, | |
| "loss": 1.71, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "grad_norm": 1.507797122001648, | |
| "learning_rate": 0.00032407407407407406, | |
| "loss": 1.7119, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "eval_accuracy": 0.34081346423562414, | |
| "eval_loss": 1.6309115886688232, | |
| "eval_runtime": 295.3653, | |
| "eval_samples_per_second": 4.828, | |
| "eval_steps_per_second": 0.305, | |
| "step": 511 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "grad_norm": 1.939512848854065, | |
| "learning_rate": 0.00030864197530864197, | |
| "loss": 1.6486, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "grad_norm": 2.4698586463928223, | |
| "learning_rate": 0.00029320987654320993, | |
| "loss": 1.6754, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "grad_norm": 1.9678858518600464, | |
| "learning_rate": 0.0002777777777777778, | |
| "loss": 1.6025, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "grad_norm": 3.6356217861175537, | |
| "learning_rate": 0.0002623456790123457, | |
| "loss": 1.6792, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "grad_norm": 2.781039237976074, | |
| "learning_rate": 0.0002469135802469136, | |
| "loss": 1.7165, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "grad_norm": 4.021714687347412, | |
| "learning_rate": 0.0002314814814814815, | |
| "loss": 1.6836, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "grad_norm": 4.392849922180176, | |
| "learning_rate": 0.00021604938271604937, | |
| "loss": 1.6981, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "eval_accuracy": 0.3288920056100982, | |
| "eval_loss": 1.6518133878707886, | |
| "eval_runtime": 319.5699, | |
| "eval_samples_per_second": 4.462, | |
| "eval_steps_per_second": 0.282, | |
| "step": 584 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "grad_norm": 2.291691303253174, | |
| "learning_rate": 0.0002006172839506173, | |
| "loss": 1.6548, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "grad_norm": 4.06191873550415, | |
| "learning_rate": 0.00018518518518518518, | |
| "loss": 1.5957, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "grad_norm": 3.735381603240967, | |
| "learning_rate": 0.0001697530864197531, | |
| "loss": 1.6757, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "grad_norm": 3.7696151733398438, | |
| "learning_rate": 0.00015432098765432098, | |
| "loss": 1.6349, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "grad_norm": 2.210860013961792, | |
| "learning_rate": 0.0001388888888888889, | |
| "loss": 1.7561, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 8.08, | |
| "grad_norm": 2.812994956970215, | |
| "learning_rate": 0.0001234567901234568, | |
| "loss": 1.6331, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 8.09, | |
| "grad_norm": 3.606325626373291, | |
| "learning_rate": 0.00010802469135802468, | |
| "loss": 1.7066, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "eval_accuracy": 0.33099579242636745, | |
| "eval_loss": 1.6313475370407104, | |
| "eval_runtime": 304.3275, | |
| "eval_samples_per_second": 4.686, | |
| "eval_steps_per_second": 0.296, | |
| "step": 657 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 2.993828296661377, | |
| "learning_rate": 9.259259259259259e-05, | |
| "loss": 1.6485, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "grad_norm": 1.9592925310134888, | |
| "learning_rate": 7.716049382716049e-05, | |
| "loss": 1.6776, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "grad_norm": 3.105025291442871, | |
| "learning_rate": 6.17283950617284e-05, | |
| "loss": 1.6466, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "grad_norm": 3.643643856048584, | |
| "learning_rate": 4.6296296296296294e-05, | |
| "loss": 1.6544, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "grad_norm": 2.5237057209014893, | |
| "learning_rate": 3.08641975308642e-05, | |
| "loss": 1.6616, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "grad_norm": 3.834527015686035, | |
| "learning_rate": 1.54320987654321e-05, | |
| "loss": 1.6271, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "grad_norm": 1.777999997138977, | |
| "learning_rate": 0.0, | |
| "loss": 1.6476, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "eval_accuracy": 0.3288920056100982, | |
| "eval_loss": 1.6337770223617554, | |
| "eval_runtime": 340.2408, | |
| "eval_samples_per_second": 4.191, | |
| "eval_steps_per_second": 0.265, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "step": 720, | |
| "total_flos": 1.4231811343419113e+19, | |
| "train_loss": 1.6995894723468357, | |
| "train_runtime": 6776.3985, | |
| "train_samples_per_second": 1.7, | |
| "train_steps_per_second": 0.106 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "eval_accuracy": 0.34811715481171546, | |
| "eval_loss": 1.703281283378601, | |
| "eval_runtime": 256.932, | |
| "eval_samples_per_second": 4.651, | |
| "eval_steps_per_second": 0.292, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "eval_accuracy": 0.34811715481171546, | |
| "eval_loss": 1.7031110525131226, | |
| "eval_runtime": 266.484, | |
| "eval_samples_per_second": 4.484, | |
| "eval_steps_per_second": 0.281, | |
| "step": 720 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 720, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 1.4231811343419113e+19, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |