| { | |
| "best_metric": 0.9642160052049447, | |
| "best_model_checkpoint": "videomae-base-finetuned-ElderReact-Fear/checkpoint-77", | |
| "epoch": 4.189473684210526, | |
| "eval_steps": 500, | |
| "global_step": 380, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 6.242678642272949, | |
| "learning_rate": 1.3157894736842106e-05, | |
| "loss": 0.5133, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 4.866024971008301, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.4234, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 6.626444339752197, | |
| "learning_rate": 3.9473684210526316e-05, | |
| "loss": 0.5058, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 3.6388182640075684, | |
| "learning_rate": 4.970760233918128e-05, | |
| "loss": 0.4791, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 2.3752734661102295, | |
| "learning_rate": 4.824561403508772e-05, | |
| "loss": 0.3841, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 2.3690688610076904, | |
| "learning_rate": 4.678362573099415e-05, | |
| "loss": 0.4903, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 2.5726335048675537, | |
| "learning_rate": 4.5321637426900585e-05, | |
| "loss": 0.2935, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_f1": 0.9642160052049447, | |
| "eval_loss": 0.25238940119743347, | |
| "eval_runtime": 1187.3545, | |
| "eval_samples_per_second": 0.67, | |
| "eval_steps_per_second": 0.084, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 3.456474542617798, | |
| "learning_rate": 4.3859649122807014e-05, | |
| "loss": 0.4967, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 1.242908239364624, | |
| "learning_rate": 4.239766081871345e-05, | |
| "loss": 0.4073, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 3.6170222759246826, | |
| "learning_rate": 4.093567251461988e-05, | |
| "loss": 0.5547, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "grad_norm": 2.8330729007720947, | |
| "learning_rate": 3.9473684210526316e-05, | |
| "loss": 0.363, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "grad_norm": 2.8961918354034424, | |
| "learning_rate": 3.8011695906432746e-05, | |
| "loss": 0.4056, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "grad_norm": 5.695966720581055, | |
| "learning_rate": 3.654970760233918e-05, | |
| "loss": 0.3925, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "grad_norm": 3.3888511657714844, | |
| "learning_rate": 3.508771929824561e-05, | |
| "loss": 0.4257, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "grad_norm": 7.591073513031006, | |
| "learning_rate": 3.362573099415205e-05, | |
| "loss": 0.3966, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_f1": 0.9642160052049447, | |
| "eval_loss": 0.25368985533714294, | |
| "eval_runtime": 1229.9142, | |
| "eval_samples_per_second": 0.647, | |
| "eval_steps_per_second": 0.081, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 3.995485305786133, | |
| "learning_rate": 3.216374269005848e-05, | |
| "loss": 0.5168, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 2.2910969257354736, | |
| "learning_rate": 3.0701754385964913e-05, | |
| "loss": 0.4298, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "grad_norm": 5.8856964111328125, | |
| "learning_rate": 2.9239766081871346e-05, | |
| "loss": 0.4855, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "grad_norm": 1.2029250860214233, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.2937, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "grad_norm": 1.0417933464050293, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.2089, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 0.3655168116092682, | |
| "learning_rate": 2.485380116959064e-05, | |
| "loss": 0.5043, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "grad_norm": 2.3576948642730713, | |
| "learning_rate": 2.3391812865497074e-05, | |
| "loss": 0.2995, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 2.619518280029297, | |
| "learning_rate": 2.1929824561403507e-05, | |
| "loss": 0.3792, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "eval_f1": 0.9642160052049447, | |
| "eval_loss": 0.2470169961452484, | |
| "eval_runtime": 1353.2793, | |
| "eval_samples_per_second": 0.588, | |
| "eval_steps_per_second": 0.074, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "grad_norm": 3.8343122005462646, | |
| "learning_rate": 2.046783625730994e-05, | |
| "loss": 0.3818, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "grad_norm": 3.7524924278259277, | |
| "learning_rate": 1.9005847953216373e-05, | |
| "loss": 0.3958, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "grad_norm": 2.071563720703125, | |
| "learning_rate": 1.7543859649122806e-05, | |
| "loss": 0.3903, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 2.1228742599487305, | |
| "learning_rate": 1.608187134502924e-05, | |
| "loss": 0.4532, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "grad_norm": 2.22047758102417, | |
| "learning_rate": 1.4619883040935673e-05, | |
| "loss": 0.3909, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "grad_norm": 1.4722857475280762, | |
| "learning_rate": 1.3157894736842106e-05, | |
| "loss": 0.44, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "grad_norm": 1.8657927513122559, | |
| "learning_rate": 1.1695906432748537e-05, | |
| "loss": 0.3666, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "eval_f1": 0.9642160052049447, | |
| "eval_loss": 0.2569561004638672, | |
| "eval_runtime": 952.0022, | |
| "eval_samples_per_second": 0.836, | |
| "eval_steps_per_second": 0.105, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "grad_norm": 6.351089000701904, | |
| "learning_rate": 1.023391812865497e-05, | |
| "loss": 0.3108, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "grad_norm": 2.317436933517456, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 0.2831, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "grad_norm": 2.794163703918457, | |
| "learning_rate": 7.3099415204678366e-06, | |
| "loss": 0.2504, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "grad_norm": 2.761640787124634, | |
| "learning_rate": 5.8479532163742686e-06, | |
| "loss": 0.4861, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "grad_norm": 4.870150089263916, | |
| "learning_rate": 4.3859649122807014e-06, | |
| "loss": 0.4329, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "grad_norm": 4.540055751800537, | |
| "learning_rate": 2.9239766081871343e-06, | |
| "loss": 0.3709, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "grad_norm": 4.406496524810791, | |
| "learning_rate": 1.4619883040935671e-06, | |
| "loss": 0.2996, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "grad_norm": 3.185987949371338, | |
| "learning_rate": 0.0, | |
| "loss": 0.2803, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "eval_f1": 0.9642160052049447, | |
| "eval_loss": 0.24840089678764343, | |
| "eval_runtime": 680.4356, | |
| "eval_samples_per_second": 1.17, | |
| "eval_steps_per_second": 0.147, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "step": 380, | |
| "total_flos": 3.783052093636215e+18, | |
| "train_loss": 0.3995331350125765, | |
| "train_runtime": 14411.9404, | |
| "train_samples_per_second": 0.211, | |
| "train_steps_per_second": 0.026 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "eval_f1": 0.9517819706498952, | |
| "eval_loss": 0.3077032268047333, | |
| "eval_runtime": 515.1776, | |
| "eval_samples_per_second": 1.456, | |
| "eval_steps_per_second": 0.182, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "eval_f1": 0.9517819706498952, | |
| "eval_loss": 0.3077031970024109, | |
| "eval_runtime": 529.5496, | |
| "eval_samples_per_second": 1.416, | |
| "eval_steps_per_second": 0.178, | |
| "step": 380 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 380, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 3.783052093636215e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |