| { | |
| "best_metric": 0.33520336605890605, | |
| "best_model_checkpoint": "videomae-base-finetuned-crema-d8-finetuned-elder-creama-d-pretuned/checkpoint-145", | |
| "epoch": 9.09375, | |
| "eval_steps": 500, | |
| "global_step": 1440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 35.67122268676758, | |
| "learning_rate": 3.4722222222222224e-06, | |
| "loss": 4.1618, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 33.357269287109375, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 3.4709, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 25.177465438842773, | |
| "learning_rate": 1.0416666666666668e-05, | |
| "loss": 2.5591, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 16.02618408203125, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 2.3848, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 15.27375602722168, | |
| "learning_rate": 1.736111111111111e-05, | |
| "loss": 2.1806, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 11.085648536682129, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 1.983, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 8.703251838684082, | |
| "learning_rate": 2.4305555555555558e-05, | |
| "loss": 1.8467, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 8.859769821166992, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 1.7654, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 8.25590705871582, | |
| "learning_rate": 3.125e-05, | |
| "loss": 1.6852, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 7.006203651428223, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 1.7227, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 6.808337688446045, | |
| "learning_rate": 3.8194444444444444e-05, | |
| "loss": 1.8047, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 7.12739372253418, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.711, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 6.718637466430664, | |
| "learning_rate": 4.5138888888888894e-05, | |
| "loss": 1.7888, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 5.254705905914307, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 1.6909, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_accuracy": 0.33520336605890605, | |
| "eval_loss": 1.6483688354492188, | |
| "eval_runtime": 1067.0455, | |
| "eval_samples_per_second": 1.336, | |
| "eval_steps_per_second": 0.168, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 6.445611476898193, | |
| "learning_rate": 4.976851851851852e-05, | |
| "loss": 1.5898, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 7.012285232543945, | |
| "learning_rate": 4.938271604938271e-05, | |
| "loss": 1.7571, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 6.888933181762695, | |
| "learning_rate": 4.899691358024692e-05, | |
| "loss": 1.7097, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 5.492002964019775, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 1.8516, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 10.119989395141602, | |
| "learning_rate": 4.8225308641975306e-05, | |
| "loss": 1.7601, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 6.299036026000977, | |
| "learning_rate": 4.783950617283951e-05, | |
| "loss": 1.6907, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 5.24801778793335, | |
| "learning_rate": 4.745370370370371e-05, | |
| "loss": 1.6796, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 4.786376476287842, | |
| "learning_rate": 4.70679012345679e-05, | |
| "loss": 1.6915, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 6.203883647918701, | |
| "learning_rate": 4.66820987654321e-05, | |
| "loss": 1.6441, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 5.037710189819336, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 1.6633, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 5.640565872192383, | |
| "learning_rate": 4.591049382716049e-05, | |
| "loss": 1.5735, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 4.959427833557129, | |
| "learning_rate": 4.5524691358024696e-05, | |
| "loss": 1.5952, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "grad_norm": 4.178658485412598, | |
| "learning_rate": 4.5138888888888894e-05, | |
| "loss": 1.6431, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "grad_norm": 6.911637306213379, | |
| "learning_rate": 4.4753086419753084e-05, | |
| "loss": 1.6024, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 6.483658790588379, | |
| "learning_rate": 4.436728395061729e-05, | |
| "loss": 1.5944, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_accuracy": 0.27980364656381485, | |
| "eval_loss": 1.6735395193099976, | |
| "eval_runtime": 1075.3628, | |
| "eval_samples_per_second": 1.326, | |
| "eval_steps_per_second": 0.166, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "grad_norm": 6.275816917419434, | |
| "learning_rate": 4.3981481481481486e-05, | |
| "loss": 1.5577, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "grad_norm": 3.8816051483154297, | |
| "learning_rate": 4.359567901234568e-05, | |
| "loss": 1.8176, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 5.7805376052856445, | |
| "learning_rate": 4.3209876543209875e-05, | |
| "loss": 1.5956, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 7.336784839630127, | |
| "learning_rate": 4.282407407407408e-05, | |
| "loss": 1.5103, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 7.558741569519043, | |
| "learning_rate": 4.243827160493827e-05, | |
| "loss": 1.6603, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 7.094559192657471, | |
| "learning_rate": 4.205246913580247e-05, | |
| "loss": 1.6139, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 5.095736503601074, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.6272, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "grad_norm": 6.24078893661499, | |
| "learning_rate": 4.128086419753087e-05, | |
| "loss": 1.7125, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "grad_norm": 6.2533392906188965, | |
| "learning_rate": 4.089506172839506e-05, | |
| "loss": 1.4988, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "grad_norm": 4.310216426849365, | |
| "learning_rate": 4.0509259259259265e-05, | |
| "loss": 1.5615, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 5.804831504821777, | |
| "learning_rate": 4.012345679012346e-05, | |
| "loss": 1.5706, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 4.093357563018799, | |
| "learning_rate": 3.973765432098765e-05, | |
| "loss": 1.5732, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "grad_norm": 9.556779861450195, | |
| "learning_rate": 3.935185185185186e-05, | |
| "loss": 1.8022, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 4.3243279457092285, | |
| "learning_rate": 3.8966049382716055e-05, | |
| "loss": 1.5776, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "eval_accuracy": 0.3211781206171108, | |
| "eval_loss": 1.6654304265975952, | |
| "eval_runtime": 1069.7784, | |
| "eval_samples_per_second": 1.333, | |
| "eval_steps_per_second": 0.167, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 10.815625190734863, | |
| "learning_rate": 3.8580246913580246e-05, | |
| "loss": 1.6887, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "grad_norm": 7.0304670333862305, | |
| "learning_rate": 3.8194444444444444e-05, | |
| "loss": 1.5767, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "grad_norm": 13.4127836227417, | |
| "learning_rate": 3.780864197530865e-05, | |
| "loss": 1.6149, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "grad_norm": 5.7433881759643555, | |
| "learning_rate": 3.742283950617284e-05, | |
| "loss": 1.5196, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "grad_norm": 3.832094430923462, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 1.626, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "grad_norm": 4.1515793800354, | |
| "learning_rate": 3.665123456790124e-05, | |
| "loss": 1.4536, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "grad_norm": 9.092682838439941, | |
| "learning_rate": 3.626543209876543e-05, | |
| "loss": 1.4983, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "grad_norm": 10.133805274963379, | |
| "learning_rate": 3.587962962962963e-05, | |
| "loss": 1.6459, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "grad_norm": 5.816834926605225, | |
| "learning_rate": 3.5493827160493834e-05, | |
| "loss": 1.4871, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "grad_norm": 5.276829242706299, | |
| "learning_rate": 3.5108024691358025e-05, | |
| "loss": 1.696, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "grad_norm": 5.181794166564941, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 1.5963, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "grad_norm": 5.877143859863281, | |
| "learning_rate": 3.4336419753086427e-05, | |
| "loss": 1.6116, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "grad_norm": 3.9388859272003174, | |
| "learning_rate": 3.395061728395062e-05, | |
| "loss": 1.5434, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "grad_norm": 8.216493606567383, | |
| "learning_rate": 3.3564814814814815e-05, | |
| "loss": 1.7139, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 5.945636749267578, | |
| "learning_rate": 3.317901234567901e-05, | |
| "loss": 1.6768, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "eval_accuracy": 0.18583450210378682, | |
| "eval_loss": 1.7330094575881958, | |
| "eval_runtime": 1081.1675, | |
| "eval_samples_per_second": 1.319, | |
| "eval_steps_per_second": 0.166, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "grad_norm": 5.201643943786621, | |
| "learning_rate": 3.279320987654321e-05, | |
| "loss": 1.6145, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "grad_norm": 5.568819522857666, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 1.5311, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "grad_norm": 4.623745441436768, | |
| "learning_rate": 3.2021604938271605e-05, | |
| "loss": 1.5345, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "grad_norm": 7.134936809539795, | |
| "learning_rate": 3.16358024691358e-05, | |
| "loss": 1.6203, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "grad_norm": 7.0437469482421875, | |
| "learning_rate": 3.125e-05, | |
| "loss": 1.4597, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "grad_norm": 6.131565570831299, | |
| "learning_rate": 3.08641975308642e-05, | |
| "loss": 1.5898, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "grad_norm": 6.676046371459961, | |
| "learning_rate": 3.04783950617284e-05, | |
| "loss": 1.5425, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "grad_norm": 5.616217136383057, | |
| "learning_rate": 3.0092592592592593e-05, | |
| "loss": 1.4983, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "grad_norm": 8.613158226013184, | |
| "learning_rate": 2.970679012345679e-05, | |
| "loss": 1.6394, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "grad_norm": 7.01395320892334, | |
| "learning_rate": 2.9320987654320992e-05, | |
| "loss": 1.6682, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "grad_norm": 6.950262546539307, | |
| "learning_rate": 2.8935185185185186e-05, | |
| "loss": 1.5914, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "grad_norm": 7.496145248413086, | |
| "learning_rate": 2.8549382716049384e-05, | |
| "loss": 1.5062, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "grad_norm": 7.298316478729248, | |
| "learning_rate": 2.8163580246913578e-05, | |
| "loss": 1.5272, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "grad_norm": 8.239325523376465, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 1.6108, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "eval_accuracy": 0.2503506311360449, | |
| "eval_loss": 1.691924810409546, | |
| "eval_runtime": 1114.3234, | |
| "eval_samples_per_second": 1.28, | |
| "eval_steps_per_second": 0.161, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 5.825406551361084, | |
| "learning_rate": 2.7391975308641977e-05, | |
| "loss": 1.5194, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "grad_norm": 5.623870849609375, | |
| "learning_rate": 2.700617283950617e-05, | |
| "loss": 1.5299, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "grad_norm": 4.288653373718262, | |
| "learning_rate": 2.6620370370370372e-05, | |
| "loss": 1.6937, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "grad_norm": 7.575453758239746, | |
| "learning_rate": 2.623456790123457e-05, | |
| "loss": 1.5297, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "grad_norm": 5.92163610458374, | |
| "learning_rate": 2.5848765432098764e-05, | |
| "loss": 1.5652, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "grad_norm": 7.6441426277160645, | |
| "learning_rate": 2.5462962962962965e-05, | |
| "loss": 1.5775, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "grad_norm": 4.205317974090576, | |
| "learning_rate": 2.5077160493827162e-05, | |
| "loss": 1.6554, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "grad_norm": 7.7730512619018555, | |
| "learning_rate": 2.4691358024691357e-05, | |
| "loss": 1.5925, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "grad_norm": 5.722338676452637, | |
| "learning_rate": 2.4305555555555558e-05, | |
| "loss": 1.5774, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "grad_norm": 8.397584915161133, | |
| "learning_rate": 2.3919753086419755e-05, | |
| "loss": 1.6021, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "grad_norm": 8.244030952453613, | |
| "learning_rate": 2.353395061728395e-05, | |
| "loss": 1.4784, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "grad_norm": 5.344796180725098, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 1.4185, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "grad_norm": 6.568389892578125, | |
| "learning_rate": 2.2762345679012348e-05, | |
| "loss": 1.5829, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "grad_norm": 4.663444995880127, | |
| "learning_rate": 2.2376543209876542e-05, | |
| "loss": 1.529, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "grad_norm": 4.336330890655518, | |
| "learning_rate": 2.1990740740740743e-05, | |
| "loss": 1.5103, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "eval_accuracy": 0.2805049088359046, | |
| "eval_loss": 1.6524486541748047, | |
| "eval_runtime": 1069.6905, | |
| "eval_samples_per_second": 1.333, | |
| "eval_steps_per_second": 0.167, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "grad_norm": 4.447099685668945, | |
| "learning_rate": 2.1604938271604937e-05, | |
| "loss": 1.4805, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "grad_norm": 4.537060260772705, | |
| "learning_rate": 2.1219135802469135e-05, | |
| "loss": 1.5277, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "grad_norm": 6.641722202301025, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 1.3914, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "grad_norm": 11.075356483459473, | |
| "learning_rate": 2.044753086419753e-05, | |
| "loss": 1.5228, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "grad_norm": 7.371598720550537, | |
| "learning_rate": 2.006172839506173e-05, | |
| "loss": 1.6399, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "grad_norm": 7.182305812835693, | |
| "learning_rate": 1.967592592592593e-05, | |
| "loss": 1.5727, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "grad_norm": 6.9517598152160645, | |
| "learning_rate": 1.9290123456790123e-05, | |
| "loss": 1.5919, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "grad_norm": 5.393679618835449, | |
| "learning_rate": 1.8904320987654324e-05, | |
| "loss": 1.5038, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "grad_norm": 6.600991249084473, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 1.5993, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "grad_norm": 5.317991733551025, | |
| "learning_rate": 1.8132716049382716e-05, | |
| "loss": 1.5051, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "grad_norm": 6.535513401031494, | |
| "learning_rate": 1.7746913580246917e-05, | |
| "loss": 1.5216, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "grad_norm": 7.035238265991211, | |
| "learning_rate": 1.736111111111111e-05, | |
| "loss": 1.508, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "grad_norm": 5.729631423950195, | |
| "learning_rate": 1.697530864197531e-05, | |
| "loss": 1.6134, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "grad_norm": 5.01363468170166, | |
| "learning_rate": 1.6589506172839506e-05, | |
| "loss": 1.5447, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "eval_accuracy": 0.3085553997194951, | |
| "eval_loss": 1.6767175197601318, | |
| "eval_runtime": 993.337, | |
| "eval_samples_per_second": 1.436, | |
| "eval_steps_per_second": 0.18, | |
| "step": 1015 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 5.760792255401611, | |
| "learning_rate": 1.6203703703703704e-05, | |
| "loss": 1.5299, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "grad_norm": 3.973865032196045, | |
| "learning_rate": 1.58179012345679e-05, | |
| "loss": 1.3628, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "grad_norm": 4.775136947631836, | |
| "learning_rate": 1.54320987654321e-05, | |
| "loss": 1.5003, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "grad_norm": 5.997446060180664, | |
| "learning_rate": 1.5046296296296297e-05, | |
| "loss": 1.4961, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "grad_norm": 6.616335391998291, | |
| "learning_rate": 1.4660493827160496e-05, | |
| "loss": 1.4256, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "grad_norm": 5.351243495941162, | |
| "learning_rate": 1.4274691358024692e-05, | |
| "loss": 1.4124, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "grad_norm": 7.641751289367676, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 1.4978, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "grad_norm": 8.547698020935059, | |
| "learning_rate": 1.3503086419753085e-05, | |
| "loss": 1.4808, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "grad_norm": 7.036251068115234, | |
| "learning_rate": 1.3117283950617285e-05, | |
| "loss": 1.4367, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "grad_norm": 5.7725982666015625, | |
| "learning_rate": 1.2731481481481482e-05, | |
| "loss": 1.4533, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "grad_norm": 6.249902725219727, | |
| "learning_rate": 1.2345679012345678e-05, | |
| "loss": 1.5726, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "grad_norm": 9.042238235473633, | |
| "learning_rate": 1.1959876543209878e-05, | |
| "loss": 1.47, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "grad_norm": 8.399900436401367, | |
| "learning_rate": 1.1574074074074075e-05, | |
| "loss": 1.5411, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "grad_norm": 6.547450065612793, | |
| "learning_rate": 1.1188271604938271e-05, | |
| "loss": 1.5962, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "grad_norm": 10.586703300476074, | |
| "learning_rate": 1.0802469135802469e-05, | |
| "loss": 1.5237, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "eval_accuracy": 0.2552594670406732, | |
| "eval_loss": 1.7328603267669678, | |
| "eval_runtime": 1099.3691, | |
| "eval_samples_per_second": 1.297, | |
| "eval_steps_per_second": 0.163, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "grad_norm": 6.408758163452148, | |
| "learning_rate": 1.0416666666666668e-05, | |
| "loss": 1.2435, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "grad_norm": 6.371536731719971, | |
| "learning_rate": 1.0030864197530866e-05, | |
| "loss": 1.3234, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "grad_norm": 5.1275410652160645, | |
| "learning_rate": 9.645061728395062e-06, | |
| "loss": 1.4336, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "grad_norm": 6.502357006072998, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 1.425, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "grad_norm": 6.920971870422363, | |
| "learning_rate": 8.873456790123458e-06, | |
| "loss": 1.462, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "grad_norm": 6.478498935699463, | |
| "learning_rate": 8.487654320987654e-06, | |
| "loss": 1.4694, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "grad_norm": 8.00242805480957, | |
| "learning_rate": 8.101851851851852e-06, | |
| "loss": 1.5305, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "grad_norm": 5.735220909118652, | |
| "learning_rate": 7.71604938271605e-06, | |
| "loss": 1.5784, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "grad_norm": 7.882507801055908, | |
| "learning_rate": 7.330246913580248e-06, | |
| "loss": 1.6219, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "grad_norm": 3.4598042964935303, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 1.4463, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 8.08, | |
| "grad_norm": 6.024099349975586, | |
| "learning_rate": 6.558641975308642e-06, | |
| "loss": 1.3849, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 8.08, | |
| "grad_norm": 5.08905029296875, | |
| "learning_rate": 6.172839506172839e-06, | |
| "loss": 1.507, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 8.09, | |
| "grad_norm": 5.94057559967041, | |
| "learning_rate": 5.787037037037038e-06, | |
| "loss": 1.5478, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "grad_norm": 7.8740458488464355, | |
| "learning_rate": 5.401234567901234e-06, | |
| "loss": 1.4397, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "eval_accuracy": 0.24754558204768584, | |
| "eval_loss": 1.7292535305023193, | |
| "eval_runtime": 1078.3495, | |
| "eval_samples_per_second": 1.322, | |
| "eval_steps_per_second": 0.166, | |
| "step": 1305 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 6.3548455238342285, | |
| "learning_rate": 5.015432098765433e-06, | |
| "loss": 1.3801, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "grad_norm": 6.385293960571289, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 1.57, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "grad_norm": 5.050124168395996, | |
| "learning_rate": 4.243827160493827e-06, | |
| "loss": 1.3412, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "grad_norm": 8.06653118133545, | |
| "learning_rate": 3.858024691358025e-06, | |
| "loss": 1.3923, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "grad_norm": 6.628742694854736, | |
| "learning_rate": 3.4722222222222224e-06, | |
| "loss": 1.4682, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "grad_norm": 5.726457118988037, | |
| "learning_rate": 3.0864197530864196e-06, | |
| "loss": 1.4295, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "grad_norm": 6.105814456939697, | |
| "learning_rate": 2.700617283950617e-06, | |
| "loss": 1.3765, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "grad_norm": 8.145650863647461, | |
| "learning_rate": 2.3148148148148148e-06, | |
| "loss": 1.3884, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "grad_norm": 5.698288440704346, | |
| "learning_rate": 1.9290123456790124e-06, | |
| "loss": 1.501, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "grad_norm": 7.972877025604248, | |
| "learning_rate": 1.5432098765432098e-06, | |
| "loss": 1.4569, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "grad_norm": 5.555938243865967, | |
| "learning_rate": 1.1574074074074074e-06, | |
| "loss": 1.4661, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "grad_norm": 6.897174835205078, | |
| "learning_rate": 7.716049382716049e-07, | |
| "loss": 1.3853, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "grad_norm": 6.399707794189453, | |
| "learning_rate": 3.8580246913580245e-07, | |
| "loss": 1.4631, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "grad_norm": 4.795849800109863, | |
| "learning_rate": 0.0, | |
| "loss": 1.4544, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "eval_accuracy": 0.2517531556802244, | |
| "eval_loss": 1.7367732524871826, | |
| "eval_runtime": 1094.539, | |
| "eval_samples_per_second": 1.303, | |
| "eval_steps_per_second": 0.164, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "step": 1440, | |
| "total_flos": 1.4321531194283852e+19, | |
| "train_loss": 1.6113481746779548, | |
| "train_runtime": 27059.6186, | |
| "train_samples_per_second": 0.426, | |
| "train_steps_per_second": 0.053 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "eval_accuracy": 0.3389121338912134, | |
| "eval_loss": 1.6702075004577637, | |
| "eval_runtime": 905.0071, | |
| "eval_samples_per_second": 1.32, | |
| "eval_steps_per_second": 0.166, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "eval_accuracy": 0.3389121338912134, | |
| "eval_loss": 1.6702075004577637, | |
| "eval_runtime": 908.3161, | |
| "eval_samples_per_second": 1.316, | |
| "eval_steps_per_second": 0.165, | |
| "step": 1440 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 1.4321531194283852e+19, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |