| { | |
| "best_metric": 0.23411221908022026, | |
| "best_model_checkpoint": "videomae-base-finetuned-ucfcrime-full2/checkpoint-616", | |
| "epoch": 7.12, | |
| "eval_steps": 500, | |
| "global_step": 700, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 9.479973793029785, | |
| "learning_rate": 7.142857142857143e-06, | |
| "loss": 2.6626, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 9.114124298095703, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 2.6526, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 8.465902328491211, | |
| "learning_rate": 2.1428571428571428e-05, | |
| "loss": 2.6297, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 8.733708381652832, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 2.5219, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 7.121254920959473, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 2.5271, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 7.228466510772705, | |
| "learning_rate": 4.2857142857142856e-05, | |
| "loss": 2.5257, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 7.514686584472656, | |
| "learning_rate": 5e-05, | |
| "loss": 2.5942, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 8.028372764587402, | |
| "learning_rate": 4.9206349206349204e-05, | |
| "loss": 2.5836, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_accuracy": 0.20799226075308827, | |
| "eval_loss": 2.494396924972534, | |
| "eval_runtime": 2684.2234, | |
| "eval_samples_per_second": 5.006, | |
| "eval_steps_per_second": 0.626, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 9.068702697753906, | |
| "learning_rate": 4.841269841269841e-05, | |
| "loss": 2.5532, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 9.19058609008789, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 2.4752, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 6.300441741943359, | |
| "learning_rate": 4.682539682539683e-05, | |
| "loss": 2.512, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 6.24762487411499, | |
| "learning_rate": 4.603174603174603e-05, | |
| "loss": 2.6188, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 6.59229850769043, | |
| "learning_rate": 4.523809523809524e-05, | |
| "loss": 2.3887, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 7.241441249847412, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 2.4597, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "grad_norm": 6.4515275955200195, | |
| "learning_rate": 4.3650793650793655e-05, | |
| "loss": 2.4792, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 6.551385879516602, | |
| "learning_rate": 4.2857142857142856e-05, | |
| "loss": 2.4636, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 5.237429618835449, | |
| "learning_rate": 4.2063492063492065e-05, | |
| "loss": 2.3212, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "eval_accuracy": 0.17725852061318648, | |
| "eval_loss": 2.5854506492614746, | |
| "eval_runtime": 2671.3135, | |
| "eval_samples_per_second": 5.03, | |
| "eval_steps_per_second": 0.629, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "grad_norm": 8.875786781311035, | |
| "learning_rate": 4.126984126984127e-05, | |
| "loss": 2.3138, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 7.928117275238037, | |
| "learning_rate": 4.047619047619048e-05, | |
| "loss": 2.535, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 7.844710350036621, | |
| "learning_rate": 3.968253968253968e-05, | |
| "loss": 2.4373, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 7.789624214172363, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 2.3984, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "grad_norm": 7.342363357543945, | |
| "learning_rate": 3.809523809523809e-05, | |
| "loss": 2.1948, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 6.116743564605713, | |
| "learning_rate": 3.730158730158731e-05, | |
| "loss": 2.3234, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "grad_norm": 8.138971328735352, | |
| "learning_rate": 3.650793650793651e-05, | |
| "loss": 2.3012, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 7.563113689422607, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 2.4211, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "grad_norm": 8.180752754211426, | |
| "learning_rate": 3.492063492063492e-05, | |
| "loss": 2.2333, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "eval_accuracy": 0.10455424914421789, | |
| "eval_loss": 2.627028226852417, | |
| "eval_runtime": 2669.9747, | |
| "eval_samples_per_second": 5.033, | |
| "eval_steps_per_second": 0.629, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "grad_norm": 8.733146667480469, | |
| "learning_rate": 3.412698412698413e-05, | |
| "loss": 2.3931, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "grad_norm": 8.307793617248535, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 2.1948, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "grad_norm": 9.008221626281738, | |
| "learning_rate": 3.253968253968254e-05, | |
| "loss": 2.3545, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "grad_norm": 8.743054389953613, | |
| "learning_rate": 3.1746031746031745e-05, | |
| "loss": 2.1182, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "grad_norm": 7.782846450805664, | |
| "learning_rate": 3.095238095238095e-05, | |
| "loss": 2.2446, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "grad_norm": 9.437503814697266, | |
| "learning_rate": 3.0158730158730158e-05, | |
| "loss": 2.3095, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "grad_norm": 8.776910781860352, | |
| "learning_rate": 2.9365079365079366e-05, | |
| "loss": 2.2165, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "grad_norm": 7.262299060821533, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 1.9958, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": 7.0002288818359375, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 1.985, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "eval_accuracy": 0.2108944783449918, | |
| "eval_loss": 2.4058477878570557, | |
| "eval_runtime": 2676.3678, | |
| "eval_samples_per_second": 5.021, | |
| "eval_steps_per_second": 0.628, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "grad_norm": 13.075250625610352, | |
| "learning_rate": 2.6984126984126984e-05, | |
| "loss": 1.9445, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "grad_norm": 11.625497817993164, | |
| "learning_rate": 2.6190476190476192e-05, | |
| "loss": 1.9884, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "grad_norm": 9.822185516357422, | |
| "learning_rate": 2.5396825396825397e-05, | |
| "loss": 2.1677, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "grad_norm": 9.551963806152344, | |
| "learning_rate": 2.4603174603174602e-05, | |
| "loss": 2.1942, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "grad_norm": 10.00942325592041, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 2.1495, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "grad_norm": 7.805147171020508, | |
| "learning_rate": 2.3015873015873015e-05, | |
| "loss": 2.1318, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "grad_norm": 10.390979766845703, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 2.0618, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "grad_norm": 8.040055274963379, | |
| "learning_rate": 2.1428571428571428e-05, | |
| "loss": 2.1917, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "grad_norm": 19.272586822509766, | |
| "learning_rate": 2.0634920634920636e-05, | |
| "loss": 2.194, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "eval_accuracy": 0.22354517041226374, | |
| "eval_loss": 2.3653502464294434, | |
| "eval_runtime": 2678.1886, | |
| "eval_samples_per_second": 5.018, | |
| "eval_steps_per_second": 0.627, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "grad_norm": 8.98901081085205, | |
| "learning_rate": 1.984126984126984e-05, | |
| "loss": 1.9553, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "grad_norm": 9.513757705688477, | |
| "learning_rate": 1.9047619047619046e-05, | |
| "loss": 1.9827, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "grad_norm": 8.948848724365234, | |
| "learning_rate": 1.8253968253968254e-05, | |
| "loss": 1.8577, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "grad_norm": 10.43582534790039, | |
| "learning_rate": 1.746031746031746e-05, | |
| "loss": 1.8543, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "grad_norm": 10.061915397644043, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 2.0255, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "grad_norm": 9.82913875579834, | |
| "learning_rate": 1.5873015873015872e-05, | |
| "loss": 2.0622, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "grad_norm": 7.94363260269165, | |
| "learning_rate": 1.5079365079365079e-05, | |
| "loss": 1.9135, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "grad_norm": 8.4102144241333, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 1.9796, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 5.13, | |
| "eval_accuracy": 0.22354517041226374, | |
| "eval_loss": 2.2609341144561768, | |
| "eval_runtime": 2669.641, | |
| "eval_samples_per_second": 5.034, | |
| "eval_steps_per_second": 0.629, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 9.409048080444336, | |
| "learning_rate": 1.3492063492063492e-05, | |
| "loss": 1.7939, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "grad_norm": 19.078195571899414, | |
| "learning_rate": 1.2698412698412699e-05, | |
| "loss": 1.7682, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "grad_norm": 8.306180000305176, | |
| "learning_rate": 1.1904761904761905e-05, | |
| "loss": 1.8795, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "grad_norm": 6.871161460876465, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 1.8303, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "grad_norm": 11.705153465270996, | |
| "learning_rate": 1.0317460317460318e-05, | |
| "loss": 1.7997, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "grad_norm": 13.744942665100098, | |
| "learning_rate": 9.523809523809523e-06, | |
| "loss": 1.7532, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "grad_norm": 11.410266876220703, | |
| "learning_rate": 8.73015873015873e-06, | |
| "loss": 1.7416, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "grad_norm": 15.524273872375488, | |
| "learning_rate": 7.936507936507936e-06, | |
| "loss": 1.9096, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "grad_norm": 9.335684776306152, | |
| "learning_rate": 7.142857142857143e-06, | |
| "loss": 1.8786, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "eval_accuracy": 0.23411221908022026, | |
| "eval_loss": 2.2724666595458984, | |
| "eval_runtime": 2674.1892, | |
| "eval_samples_per_second": 5.025, | |
| "eval_steps_per_second": 0.628, | |
| "step": 616 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "grad_norm": 12.772497177124023, | |
| "learning_rate": 6.349206349206349e-06, | |
| "loss": 1.6914, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "grad_norm": 11.079833984375, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 1.6596, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "grad_norm": 11.924140930175781, | |
| "learning_rate": 4.7619047619047615e-06, | |
| "loss": 1.7996, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "grad_norm": 12.28601360321045, | |
| "learning_rate": 3.968253968253968e-06, | |
| "loss": 1.5564, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "grad_norm": 12.644978523254395, | |
| "learning_rate": 3.1746031746031746e-06, | |
| "loss": 1.8793, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "grad_norm": 11.849274635314941, | |
| "learning_rate": 2.3809523809523808e-06, | |
| "loss": 1.6106, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "grad_norm": 12.6458101272583, | |
| "learning_rate": 1.5873015873015873e-06, | |
| "loss": 1.9571, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "grad_norm": 12.309945106506348, | |
| "learning_rate": 7.936507936507937e-07, | |
| "loss": 1.6146, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 7.12, | |
| "grad_norm": 11.306781768798828, | |
| "learning_rate": 0.0, | |
| "loss": 1.71, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.12, | |
| "eval_accuracy": 0.22257776454829586, | |
| "eval_loss": 2.2227871417999268, | |
| "eval_runtime": 2677.8894, | |
| "eval_samples_per_second": 5.018, | |
| "eval_steps_per_second": 0.627, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.12, | |
| "step": 700, | |
| "total_flos": 6.943814924023628e+18, | |
| "train_loss": 2.1489518587929863, | |
| "train_runtime": 24272.1999, | |
| "train_samples_per_second": 0.231, | |
| "train_steps_per_second": 0.029 | |
| }, | |
| { | |
| "epoch": 7.12, | |
| "eval_accuracy": 0.225, | |
| "eval_loss": 2.50138521194458, | |
| "eval_runtime": 955.2551, | |
| "eval_samples_per_second": 5.192, | |
| "eval_steps_per_second": 0.649, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.12, | |
| "eval_accuracy": 0.225, | |
| "eval_loss": 2.501384973526001, | |
| "eval_runtime": 945.2716, | |
| "eval_samples_per_second": 5.247, | |
| "eval_steps_per_second": 0.656, | |
| "step": 700 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 700, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 6.943814924023628e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |