| { | |
| "best_metric": 0.9164021164021164, | |
| "best_model_checkpoint": "dinov2-base-fa-disabled-finetuned-har/checkpoint-830", | |
| "epoch": 9.91044776119403, | |
| "eval_steps": 500, | |
| "global_step": 830, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11940298507462686, | |
| "grad_norm": 38.431739807128906, | |
| "learning_rate": 6.024096385542169e-06, | |
| "loss": 3.0066, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.23880597014925373, | |
| "grad_norm": 39.09220504760742, | |
| "learning_rate": 1.2048192771084338e-05, | |
| "loss": 2.0974, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3582089552238806, | |
| "grad_norm": 36.01054000854492, | |
| "learning_rate": 1.8072289156626505e-05, | |
| "loss": 1.3051, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.47761194029850745, | |
| "grad_norm": 38.050384521484375, | |
| "learning_rate": 2.4096385542168677e-05, | |
| "loss": 1.0243, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5970149253731343, | |
| "grad_norm": 36.212276458740234, | |
| "learning_rate": 3.012048192771085e-05, | |
| "loss": 0.9214, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.7164179104477612, | |
| "grad_norm": 44.00043487548828, | |
| "learning_rate": 3.614457831325301e-05, | |
| "loss": 0.9179, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.835820895522388, | |
| "grad_norm": 50.121219635009766, | |
| "learning_rate": 4.2168674698795186e-05, | |
| "loss": 0.9083, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.9552238805970149, | |
| "grad_norm": 41.91326141357422, | |
| "learning_rate": 4.8192771084337354e-05, | |
| "loss": 0.8554, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.991044776119403, | |
| "eval_accuracy": 0.8322751322751323, | |
| "eval_loss": 0.5251870155334473, | |
| "eval_runtime": 36.3735, | |
| "eval_samples_per_second": 51.961, | |
| "eval_steps_per_second": 1.65, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 1.0746268656716418, | |
| "grad_norm": 30.97153663635254, | |
| "learning_rate": 4.953145917001339e-05, | |
| "loss": 0.9029, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.1940298507462686, | |
| "grad_norm": 42.186763763427734, | |
| "learning_rate": 4.886211512717537e-05, | |
| "loss": 0.8259, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.3134328358208955, | |
| "grad_norm": 34.84912109375, | |
| "learning_rate": 4.8192771084337354e-05, | |
| "loss": 0.8652, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.4328358208955223, | |
| "grad_norm": 33.463218688964844, | |
| "learning_rate": 4.7523427041499336e-05, | |
| "loss": 0.9613, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.5522388059701493, | |
| "grad_norm": 25.84391975402832, | |
| "learning_rate": 4.685408299866131e-05, | |
| "loss": 0.9065, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.671641791044776, | |
| "grad_norm": 37.509464263916016, | |
| "learning_rate": 4.61847389558233e-05, | |
| "loss": 0.7881, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.7910447761194028, | |
| "grad_norm": 21.088834762573242, | |
| "learning_rate": 4.5515394912985275e-05, | |
| "loss": 0.8766, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.9104477611940298, | |
| "grad_norm": 33.41486358642578, | |
| "learning_rate": 4.484605087014726e-05, | |
| "loss": 0.8162, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.9940298507462688, | |
| "eval_accuracy": 0.8597883597883598, | |
| "eval_loss": 0.4597439467906952, | |
| "eval_runtime": 21.2204, | |
| "eval_samples_per_second": 89.065, | |
| "eval_steps_per_second": 2.827, | |
| "step": 167 | |
| }, | |
| { | |
| "epoch": 2.029850746268657, | |
| "grad_norm": 37.96963882446289, | |
| "learning_rate": 4.417670682730924e-05, | |
| "loss": 0.7496, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.1492537313432836, | |
| "grad_norm": 28.700788497924805, | |
| "learning_rate": 4.350736278447122e-05, | |
| "loss": 0.678, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.2686567164179103, | |
| "grad_norm": 26.9548397064209, | |
| "learning_rate": 4.2838018741633203e-05, | |
| "loss": 0.7308, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.388059701492537, | |
| "grad_norm": 29.515504837036133, | |
| "learning_rate": 4.2168674698795186e-05, | |
| "loss": 0.7616, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.5074626865671643, | |
| "grad_norm": 25.0078067779541, | |
| "learning_rate": 4.149933065595716e-05, | |
| "loss": 0.7281, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.626865671641791, | |
| "grad_norm": 30.829315185546875, | |
| "learning_rate": 4.082998661311915e-05, | |
| "loss": 0.7334, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.746268656716418, | |
| "grad_norm": 18.689016342163086, | |
| "learning_rate": 4.0160642570281125e-05, | |
| "loss": 0.7141, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.8656716417910446, | |
| "grad_norm": 22.5184383392334, | |
| "learning_rate": 3.949129852744311e-05, | |
| "loss": 0.6661, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.9850746268656714, | |
| "grad_norm": 25.77634048461914, | |
| "learning_rate": 3.882195448460509e-05, | |
| "loss": 0.7303, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.997014925373134, | |
| "eval_accuracy": 0.8587301587301587, | |
| "eval_loss": 0.44030269980430603, | |
| "eval_runtime": 21.4202, | |
| "eval_samples_per_second": 88.234, | |
| "eval_steps_per_second": 2.801, | |
| "step": 251 | |
| }, | |
| { | |
| "epoch": 3.1044776119402986, | |
| "grad_norm": 19.202600479125977, | |
| "learning_rate": 3.815261044176707e-05, | |
| "loss": 0.6748, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.2238805970149254, | |
| "grad_norm": 23.559417724609375, | |
| "learning_rate": 3.748326639892905e-05, | |
| "loss": 0.6641, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.343283582089552, | |
| "grad_norm": 33.11872100830078, | |
| "learning_rate": 3.6813922356091035e-05, | |
| "loss": 0.6263, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.4626865671641793, | |
| "grad_norm": 20.86111831665039, | |
| "learning_rate": 3.614457831325301e-05, | |
| "loss": 0.6203, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.582089552238806, | |
| "grad_norm": 18.73584747314453, | |
| "learning_rate": 3.5475234270415e-05, | |
| "loss": 0.6101, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.701492537313433, | |
| "grad_norm": 30.896474838256836, | |
| "learning_rate": 3.4805890227576974e-05, | |
| "loss": 0.632, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.8208955223880596, | |
| "grad_norm": 23.907413482666016, | |
| "learning_rate": 3.413654618473896e-05, | |
| "loss": 0.6742, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.9402985074626864, | |
| "grad_norm": 24.39226722717285, | |
| "learning_rate": 3.346720214190094e-05, | |
| "loss": 0.5644, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8746031746031746, | |
| "eval_loss": 0.3922259509563446, | |
| "eval_runtime": 21.3995, | |
| "eval_samples_per_second": 88.32, | |
| "eval_steps_per_second": 2.804, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 4.059701492537314, | |
| "grad_norm": 18.170991897583008, | |
| "learning_rate": 3.279785809906292e-05, | |
| "loss": 0.5415, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 4.17910447761194, | |
| "grad_norm": 57.57135772705078, | |
| "learning_rate": 3.21285140562249e-05, | |
| "loss": 0.508, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.298507462686567, | |
| "grad_norm": 29.80289077758789, | |
| "learning_rate": 3.1459170013386885e-05, | |
| "loss": 0.5873, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.417910447761194, | |
| "grad_norm": 18.16728973388672, | |
| "learning_rate": 3.078982597054887e-05, | |
| "loss": 0.5387, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 4.537313432835821, | |
| "grad_norm": 21.491649627685547, | |
| "learning_rate": 3.012048192771085e-05, | |
| "loss": 0.5715, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.656716417910448, | |
| "grad_norm": 37.463436126708984, | |
| "learning_rate": 2.9451137884872827e-05, | |
| "loss": 0.5179, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.776119402985074, | |
| "grad_norm": 26.464433670043945, | |
| "learning_rate": 2.878179384203481e-05, | |
| "loss": 0.5064, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.895522388059701, | |
| "grad_norm": 20.331892013549805, | |
| "learning_rate": 2.8112449799196788e-05, | |
| "loss": 0.5672, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 4.991044776119403, | |
| "eval_accuracy": 0.8857142857142857, | |
| "eval_loss": 0.3783741593360901, | |
| "eval_runtime": 21.5183, | |
| "eval_samples_per_second": 87.832, | |
| "eval_steps_per_second": 2.788, | |
| "step": 418 | |
| }, | |
| { | |
| "epoch": 5.014925373134329, | |
| "grad_norm": 16.20704460144043, | |
| "learning_rate": 2.7443105756358774e-05, | |
| "loss": 0.5492, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 5.134328358208955, | |
| "grad_norm": 24.48471450805664, | |
| "learning_rate": 2.6773761713520752e-05, | |
| "loss": 0.4528, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 5.253731343283582, | |
| "grad_norm": 27.444704055786133, | |
| "learning_rate": 2.6104417670682734e-05, | |
| "loss": 0.4718, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 5.373134328358209, | |
| "grad_norm": 19.921144485473633, | |
| "learning_rate": 2.5435073627844713e-05, | |
| "loss": 0.4575, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 5.492537313432836, | |
| "grad_norm": 24.932510375976562, | |
| "learning_rate": 2.4765729585006695e-05, | |
| "loss": 0.5138, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 5.611940298507463, | |
| "grad_norm": 29.121278762817383, | |
| "learning_rate": 2.4096385542168677e-05, | |
| "loss": 0.4647, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 5.731343283582089, | |
| "grad_norm": 21.379953384399414, | |
| "learning_rate": 2.3427041499330656e-05, | |
| "loss": 0.4648, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 5.850746268656716, | |
| "grad_norm": 23.85083770751953, | |
| "learning_rate": 2.2757697456492638e-05, | |
| "loss": 0.4935, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 5.970149253731344, | |
| "grad_norm": 17.832536697387695, | |
| "learning_rate": 2.208835341365462e-05, | |
| "loss": 0.454, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.994029850746268, | |
| "eval_accuracy": 0.8830687830687831, | |
| "eval_loss": 0.38558998703956604, | |
| "eval_runtime": 21.2656, | |
| "eval_samples_per_second": 88.876, | |
| "eval_steps_per_second": 2.821, | |
| "step": 502 | |
| }, | |
| { | |
| "epoch": 6.08955223880597, | |
| "grad_norm": 33.72905349731445, | |
| "learning_rate": 2.1419009370816602e-05, | |
| "loss": 0.4187, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 6.208955223880597, | |
| "grad_norm": 20.838478088378906, | |
| "learning_rate": 2.074966532797858e-05, | |
| "loss": 0.4227, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 6.3283582089552235, | |
| "grad_norm": 21.023418426513672, | |
| "learning_rate": 2.0080321285140562e-05, | |
| "loss": 0.466, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 6.447761194029851, | |
| "grad_norm": 17.22841453552246, | |
| "learning_rate": 1.9410977242302544e-05, | |
| "loss": 0.4033, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 6.567164179104478, | |
| "grad_norm": 16.87044906616211, | |
| "learning_rate": 1.8741633199464527e-05, | |
| "loss": 0.4196, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 6.686567164179104, | |
| "grad_norm": 21.26498794555664, | |
| "learning_rate": 1.8072289156626505e-05, | |
| "loss": 0.4334, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 6.8059701492537314, | |
| "grad_norm": 18.514463424682617, | |
| "learning_rate": 1.7402945113788487e-05, | |
| "loss": 0.4471, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 6.925373134328359, | |
| "grad_norm": 21.403791427612305, | |
| "learning_rate": 1.673360107095047e-05, | |
| "loss": 0.4379, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 6.997014925373135, | |
| "eval_accuracy": 0.8888888888888888, | |
| "eval_loss": 0.35099852085113525, | |
| "eval_runtime": 21.3551, | |
| "eval_samples_per_second": 88.504, | |
| "eval_steps_per_second": 2.81, | |
| "step": 586 | |
| }, | |
| { | |
| "epoch": 7.044776119402985, | |
| "grad_norm": 17.73862648010254, | |
| "learning_rate": 1.606425702811245e-05, | |
| "loss": 0.3597, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 7.164179104477612, | |
| "grad_norm": 21.782960891723633, | |
| "learning_rate": 1.5394912985274433e-05, | |
| "loss": 0.3952, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 7.2835820895522385, | |
| "grad_norm": 14.208507537841797, | |
| "learning_rate": 1.4725568942436414e-05, | |
| "loss": 0.3769, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 7.402985074626866, | |
| "grad_norm": 15.11253833770752, | |
| "learning_rate": 1.4056224899598394e-05, | |
| "loss": 0.3558, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 7.522388059701493, | |
| "grad_norm": 17.576494216918945, | |
| "learning_rate": 1.3386880856760376e-05, | |
| "loss": 0.3477, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 7.641791044776119, | |
| "grad_norm": 19.59687042236328, | |
| "learning_rate": 1.2717536813922356e-05, | |
| "loss": 0.3656, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 7.7611940298507465, | |
| "grad_norm": 23.697837829589844, | |
| "learning_rate": 1.2048192771084338e-05, | |
| "loss": 0.3668, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 7.880597014925373, | |
| "grad_norm": 13.360515594482422, | |
| "learning_rate": 1.1378848728246319e-05, | |
| "loss": 0.3576, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 23.808975219726562, | |
| "learning_rate": 1.0709504685408301e-05, | |
| "loss": 0.3356, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9063492063492063, | |
| "eval_loss": 0.31871286034584045, | |
| "eval_runtime": 21.467, | |
| "eval_samples_per_second": 88.042, | |
| "eval_steps_per_second": 2.795, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 8.119402985074627, | |
| "grad_norm": 14.76814079284668, | |
| "learning_rate": 1.0040160642570281e-05, | |
| "loss": 0.3394, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 8.238805970149254, | |
| "grad_norm": 15.642696380615234, | |
| "learning_rate": 9.370816599732263e-06, | |
| "loss": 0.2949, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 8.35820895522388, | |
| "grad_norm": 16.03794288635254, | |
| "learning_rate": 8.701472556894244e-06, | |
| "loss": 0.3072, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 8.477611940298507, | |
| "grad_norm": 14.671366691589355, | |
| "learning_rate": 8.032128514056226e-06, | |
| "loss": 0.3508, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 8.597014925373134, | |
| "grad_norm": 14.290385246276855, | |
| "learning_rate": 7.362784471218207e-06, | |
| "loss": 0.2763, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 8.716417910447761, | |
| "grad_norm": 17.65837287902832, | |
| "learning_rate": 6.693440428380188e-06, | |
| "loss": 0.3131, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 8.835820895522389, | |
| "grad_norm": 21.202238082885742, | |
| "learning_rate": 6.024096385542169e-06, | |
| "loss": 0.3269, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 8.955223880597014, | |
| "grad_norm": 13.900936126708984, | |
| "learning_rate": 5.3547523427041504e-06, | |
| "loss": 0.2877, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 8.991044776119402, | |
| "eval_accuracy": 0.9116402116402117, | |
| "eval_loss": 0.3208979070186615, | |
| "eval_runtime": 21.2786, | |
| "eval_samples_per_second": 88.822, | |
| "eval_steps_per_second": 2.82, | |
| "step": 753 | |
| }, | |
| { | |
| "epoch": 9.074626865671641, | |
| "grad_norm": 28.610790252685547, | |
| "learning_rate": 4.685408299866132e-06, | |
| "loss": 0.2574, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 9.194029850746269, | |
| "grad_norm": 15.897197723388672, | |
| "learning_rate": 4.016064257028113e-06, | |
| "loss": 0.269, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 9.313432835820896, | |
| "grad_norm": 16.005224227905273, | |
| "learning_rate": 3.346720214190094e-06, | |
| "loss": 0.2783, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 9.432835820895523, | |
| "grad_norm": 12.884819984436035, | |
| "learning_rate": 2.6773761713520752e-06, | |
| "loss": 0.2813, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 9.552238805970148, | |
| "grad_norm": 15.747310638427734, | |
| "learning_rate": 2.0080321285140564e-06, | |
| "loss": 0.2777, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 9.671641791044776, | |
| "grad_norm": 13.63232421875, | |
| "learning_rate": 1.3386880856760376e-06, | |
| "loss": 0.2908, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 9.791044776119403, | |
| "grad_norm": 13.028614044189453, | |
| "learning_rate": 6.693440428380188e-07, | |
| "loss": 0.2894, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 9.91044776119403, | |
| "grad_norm": 12.740144729614258, | |
| "learning_rate": 0.0, | |
| "loss": 0.2717, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 9.91044776119403, | |
| "eval_accuracy": 0.9164021164021164, | |
| "eval_loss": 0.3027323782444, | |
| "eval_runtime": 22.4884, | |
| "eval_samples_per_second": 84.043, | |
| "eval_steps_per_second": 2.668, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 9.91044776119403, | |
| "step": 830, | |
| "total_flos": 1.0844463059730432e+19, | |
| "train_loss": 0.5978287294686558, | |
| "train_runtime": 2802.0121, | |
| "train_samples_per_second": 38.223, | |
| "train_steps_per_second": 0.296 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 830, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0844463059730432e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |