| { | |
| "best_metric": 1.0, | |
| "best_model_checkpoint": "videomae-base-finetuned-bekhoaxe/checkpoint-231", | |
| "epoch": 3.25, | |
| "eval_steps": 500, | |
| "global_step": 308, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 4.9310173988342285, | |
| "learning_rate": 1.6129032258064517e-05, | |
| "loss": 0.7032, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 6.229051113128662, | |
| "learning_rate": 3.2258064516129034e-05, | |
| "loss": 0.5738, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 9.958233833312988, | |
| "learning_rate": 4.8387096774193554e-05, | |
| "loss": 0.3415, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 7.560524940490723, | |
| "learning_rate": 4.837545126353791e-05, | |
| "loss": 0.4689, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 5.026527404785156, | |
| "learning_rate": 4.657039711191336e-05, | |
| "loss": 0.2069, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.358181893825531, | |
| "learning_rate": 4.4765342960288806e-05, | |
| "loss": 0.0961, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.05433480069041252, | |
| "learning_rate": 4.296028880866426e-05, | |
| "loss": 0.2172, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_accuracy": 0.9915254237288136, | |
| "eval_loss": 0.013951542787253857, | |
| "eval_runtime": 38.5113, | |
| "eval_samples_per_second": 3.064, | |
| "eval_steps_per_second": 0.779, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 1.3378506898880005, | |
| "learning_rate": 4.115523465703972e-05, | |
| "loss": 0.3363, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 0.04731794446706772, | |
| "learning_rate": 3.935018050541516e-05, | |
| "loss": 0.195, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 0.04140738397836685, | |
| "learning_rate": 3.754512635379062e-05, | |
| "loss": 0.0049, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "grad_norm": 0.08140890300273895, | |
| "learning_rate": 3.574007220216607e-05, | |
| "loss": 0.2396, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "grad_norm": 0.09178142994642258, | |
| "learning_rate": 3.3935018050541516e-05, | |
| "loss": 0.0746, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "grad_norm": 0.031925346702337265, | |
| "learning_rate": 3.212996389891697e-05, | |
| "loss": 0.3279, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.11948370933532715, | |
| "learning_rate": 3.032490974729242e-05, | |
| "loss": 0.0021, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 0.05676674470305443, | |
| "learning_rate": 2.851985559566787e-05, | |
| "loss": 0.1826, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_accuracy": 0.9830508474576272, | |
| "eval_loss": 0.042561955749988556, | |
| "eval_runtime": 38.1931, | |
| "eval_samples_per_second": 3.09, | |
| "eval_steps_per_second": 0.785, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 0.11792542040348053, | |
| "learning_rate": 2.6714801444043324e-05, | |
| "loss": 0.1442, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 0.10978111624717712, | |
| "learning_rate": 2.4909747292418774e-05, | |
| "loss": 0.0912, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 0.016077380627393723, | |
| "learning_rate": 2.3104693140794227e-05, | |
| "loss": 0.1236, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "grad_norm": 0.01301741786301136, | |
| "learning_rate": 2.1299638989169676e-05, | |
| "loss": 0.1501, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 0.026157772168517113, | |
| "learning_rate": 1.9494584837545125e-05, | |
| "loss": 0.18, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "grad_norm": 0.04890529438853264, | |
| "learning_rate": 1.768953068592058e-05, | |
| "loss": 0.0038, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "grad_norm": 0.038881637156009674, | |
| "learning_rate": 1.588447653429603e-05, | |
| "loss": 0.0019, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 0.01524051371961832, | |
| "learning_rate": 1.407942238267148e-05, | |
| "loss": 0.1427, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0019937974866479635, | |
| "eval_runtime": 38.3352, | |
| "eval_samples_per_second": 3.078, | |
| "eval_steps_per_second": 0.783, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "grad_norm": 0.01709669642150402, | |
| "learning_rate": 1.2274368231046932e-05, | |
| "loss": 0.0012, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "grad_norm": 0.1900138109922409, | |
| "learning_rate": 1.0469314079422383e-05, | |
| "loss": 0.1073, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "grad_norm": 0.06225905194878578, | |
| "learning_rate": 8.664259927797834e-06, | |
| "loss": 0.0029, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "grad_norm": 0.015592777170240879, | |
| "learning_rate": 6.859205776173286e-06, | |
| "loss": 0.0009, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "grad_norm": 0.016208168119192123, | |
| "learning_rate": 5.054151624548736e-06, | |
| "loss": 0.0036, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "grad_norm": 0.007616001646965742, | |
| "learning_rate": 3.2490974729241876e-06, | |
| "loss": 0.0008, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "grad_norm": 0.008425813168287277, | |
| "learning_rate": 1.4440433212996392e-06, | |
| "loss": 0.2853, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0012017178814858198, | |
| "eval_runtime": 38.094, | |
| "eval_samples_per_second": 3.098, | |
| "eval_steps_per_second": 0.788, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "step": 308, | |
| "total_flos": 1.5351515742291886e+18, | |
| "train_loss": 0.16999648642641577, | |
| "train_runtime": 584.5872, | |
| "train_samples_per_second": 2.107, | |
| "train_steps_per_second": 0.527 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "eval_accuracy": 0.9761904761904762, | |
| "eval_loss": 0.09526590257883072, | |
| "eval_runtime": 122.5253, | |
| "eval_samples_per_second": 1.714, | |
| "eval_steps_per_second": 0.433, | |
| "step": 308 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 308, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 1.5351515742291886e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |