hubert-base-ft-keyword-spotting / trainer_state.json
anton-l's picture
anton-l HF staff
End of training
ae2d99b
{
"best_metric": 0.9819064430714917,
"best_model_checkpoint": "hubert-base-ft-keyword-spotting/checkpoint-1995",
"epoch": 4.999373825923607,
"global_step": 1995,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 1.5e-06,
"loss": 2.4768,
"step": 10
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 2.4615,
"step": 20
},
{
"epoch": 0.08,
"learning_rate": 4.5e-06,
"loss": 2.4135,
"step": 30
},
{
"epoch": 0.1,
"learning_rate": 6e-06,
"loss": 2.3143,
"step": 40
},
{
"epoch": 0.13,
"learning_rate": 7.5e-06,
"loss": 2.1904,
"step": 50
},
{
"epoch": 0.15,
"learning_rate": 9e-06,
"loss": 2.047,
"step": 60
},
{
"epoch": 0.18,
"learning_rate": 1.05e-05,
"loss": 1.9046,
"step": 70
},
{
"epoch": 0.2,
"learning_rate": 1.2e-05,
"loss": 1.8282,
"step": 80
},
{
"epoch": 0.23,
"learning_rate": 1.3500000000000001e-05,
"loss": 1.7283,
"step": 90
},
{
"epoch": 0.25,
"learning_rate": 1.5e-05,
"loss": 1.7102,
"step": 100
},
{
"epoch": 0.28,
"learning_rate": 1.65e-05,
"loss": 1.5944,
"step": 110
},
{
"epoch": 0.3,
"learning_rate": 1.8e-05,
"loss": 1.5268,
"step": 120
},
{
"epoch": 0.33,
"learning_rate": 1.95e-05,
"loss": 1.5128,
"step": 130
},
{
"epoch": 0.35,
"learning_rate": 2.1e-05,
"loss": 1.5108,
"step": 140
},
{
"epoch": 0.38,
"learning_rate": 2.25e-05,
"loss": 1.5206,
"step": 150
},
{
"epoch": 0.4,
"learning_rate": 2.4e-05,
"loss": 1.5091,
"step": 160
},
{
"epoch": 0.43,
"learning_rate": 2.55e-05,
"loss": 1.4353,
"step": 170
},
{
"epoch": 0.45,
"learning_rate": 2.7000000000000002e-05,
"loss": 1.5367,
"step": 180
},
{
"epoch": 0.48,
"learning_rate": 2.8499999999999998e-05,
"loss": 1.5555,
"step": 190
},
{
"epoch": 0.5,
"learning_rate": 3e-05,
"loss": 1.5275,
"step": 200
},
{
"epoch": 0.53,
"learning_rate": 2.9832869080779945e-05,
"loss": 1.5864,
"step": 210
},
{
"epoch": 0.55,
"learning_rate": 2.9665738161559886e-05,
"loss": 1.55,
"step": 220
},
{
"epoch": 0.58,
"learning_rate": 2.9498607242339834e-05,
"loss": 1.4591,
"step": 230
},
{
"epoch": 0.6,
"learning_rate": 2.933147632311978e-05,
"loss": 1.4969,
"step": 240
},
{
"epoch": 0.63,
"learning_rate": 2.916434540389972e-05,
"loss": 1.4884,
"step": 250
},
{
"epoch": 0.65,
"learning_rate": 2.8997214484679665e-05,
"loss": 1.5086,
"step": 260
},
{
"epoch": 0.68,
"learning_rate": 2.8830083565459613e-05,
"loss": 1.5097,
"step": 270
},
{
"epoch": 0.7,
"learning_rate": 2.8662952646239554e-05,
"loss": 1.4455,
"step": 280
},
{
"epoch": 0.73,
"learning_rate": 2.84958217270195e-05,
"loss": 1.5155,
"step": 290
},
{
"epoch": 0.75,
"learning_rate": 2.8328690807799443e-05,
"loss": 1.3772,
"step": 300
},
{
"epoch": 0.78,
"learning_rate": 2.8161559888579388e-05,
"loss": 1.4063,
"step": 310
},
{
"epoch": 0.8,
"learning_rate": 2.7994428969359332e-05,
"loss": 1.3666,
"step": 320
},
{
"epoch": 0.83,
"learning_rate": 2.7827298050139277e-05,
"loss": 1.3803,
"step": 330
},
{
"epoch": 0.85,
"learning_rate": 2.7660167130919218e-05,
"loss": 1.3257,
"step": 340
},
{
"epoch": 0.88,
"learning_rate": 2.7493036211699166e-05,
"loss": 1.1917,
"step": 350
},
{
"epoch": 0.9,
"learning_rate": 2.732590529247911e-05,
"loss": 1.1957,
"step": 360
},
{
"epoch": 0.93,
"learning_rate": 2.7158774373259055e-05,
"loss": 1.164,
"step": 370
},
{
"epoch": 0.95,
"learning_rate": 2.6991643454038996e-05,
"loss": 1.1043,
"step": 380
},
{
"epoch": 0.98,
"learning_rate": 2.6824512534818944e-05,
"loss": 1.0422,
"step": 390
},
{
"epoch": 1.0,
"eval_accuracy": 0.6918211238599589,
"eval_loss": 0.8998974561691284,
"eval_runtime": 13.5007,
"eval_samples_per_second": 503.53,
"eval_steps_per_second": 15.777,
"step": 399
},
{
"epoch": 1.0,
"learning_rate": 2.665738161559889e-05,
"loss": 1.0979,
"step": 400
},
{
"epoch": 1.03,
"learning_rate": 2.649025069637883e-05,
"loss": 0.9165,
"step": 410
},
{
"epoch": 1.05,
"learning_rate": 2.6323119777158774e-05,
"loss": 0.9599,
"step": 420
},
{
"epoch": 1.08,
"learning_rate": 2.6155988857938722e-05,
"loss": 0.88,
"step": 430
},
{
"epoch": 1.1,
"learning_rate": 2.5988857938718663e-05,
"loss": 0.8921,
"step": 440
},
{
"epoch": 1.13,
"learning_rate": 2.5821727019498608e-05,
"loss": 0.8814,
"step": 450
},
{
"epoch": 1.15,
"learning_rate": 2.5654596100278553e-05,
"loss": 0.8321,
"step": 460
},
{
"epoch": 1.18,
"learning_rate": 2.5487465181058497e-05,
"loss": 0.7497,
"step": 470
},
{
"epoch": 1.2,
"learning_rate": 2.532033426183844e-05,
"loss": 0.7318,
"step": 480
},
{
"epoch": 1.23,
"learning_rate": 2.5153203342618386e-05,
"loss": 0.672,
"step": 490
},
{
"epoch": 1.25,
"learning_rate": 2.4986072423398327e-05,
"loss": 0.6778,
"step": 500
},
{
"epoch": 1.28,
"learning_rate": 2.4818941504178275e-05,
"loss": 0.6688,
"step": 510
},
{
"epoch": 1.3,
"learning_rate": 2.465181058495822e-05,
"loss": 0.6432,
"step": 520
},
{
"epoch": 1.33,
"learning_rate": 2.448467966573816e-05,
"loss": 0.615,
"step": 530
},
{
"epoch": 1.35,
"learning_rate": 2.4317548746518106e-05,
"loss": 0.5814,
"step": 540
},
{
"epoch": 1.38,
"learning_rate": 2.415041782729805e-05,
"loss": 0.527,
"step": 550
},
{
"epoch": 1.4,
"learning_rate": 2.3983286908077995e-05,
"loss": 0.5114,
"step": 560
},
{
"epoch": 1.43,
"learning_rate": 2.381615598885794e-05,
"loss": 0.5298,
"step": 570
},
{
"epoch": 1.45,
"learning_rate": 2.3649025069637884e-05,
"loss": 0.5075,
"step": 580
},
{
"epoch": 1.48,
"learning_rate": 2.3481894150417825e-05,
"loss": 0.4495,
"step": 590
},
{
"epoch": 1.5,
"learning_rate": 2.3314763231197773e-05,
"loss": 0.4965,
"step": 600
},
{
"epoch": 1.53,
"learning_rate": 2.3147632311977718e-05,
"loss": 0.4944,
"step": 610
},
{
"epoch": 1.55,
"learning_rate": 2.298050139275766e-05,
"loss": 0.4509,
"step": 620
},
{
"epoch": 1.58,
"learning_rate": 2.2813370473537603e-05,
"loss": 0.4434,
"step": 630
},
{
"epoch": 1.6,
"learning_rate": 2.264623955431755e-05,
"loss": 0.5139,
"step": 640
},
{
"epoch": 1.63,
"learning_rate": 2.2479108635097492e-05,
"loss": 0.4106,
"step": 650
},
{
"epoch": 1.65,
"learning_rate": 2.2311977715877437e-05,
"loss": 0.3633,
"step": 660
},
{
"epoch": 1.68,
"learning_rate": 2.214484679665738e-05,
"loss": 0.3641,
"step": 670
},
{
"epoch": 1.7,
"learning_rate": 2.1977715877437326e-05,
"loss": 0.417,
"step": 680
},
{
"epoch": 1.73,
"learning_rate": 2.181058495821727e-05,
"loss": 0.4298,
"step": 690
},
{
"epoch": 1.75,
"learning_rate": 2.1643454038997215e-05,
"loss": 0.3578,
"step": 700
},
{
"epoch": 1.78,
"learning_rate": 2.147632311977716e-05,
"loss": 0.3365,
"step": 710
},
{
"epoch": 1.8,
"learning_rate": 2.1309192200557104e-05,
"loss": 0.3343,
"step": 720
},
{
"epoch": 1.83,
"learning_rate": 2.114206128133705e-05,
"loss": 0.3471,
"step": 730
},
{
"epoch": 1.85,
"learning_rate": 2.0974930362116993e-05,
"loss": 0.335,
"step": 740
},
{
"epoch": 1.88,
"learning_rate": 2.0807799442896935e-05,
"loss": 0.344,
"step": 750
},
{
"epoch": 1.9,
"learning_rate": 2.0640668523676883e-05,
"loss": 0.3352,
"step": 760
},
{
"epoch": 1.93,
"learning_rate": 2.0473537604456827e-05,
"loss": 0.2792,
"step": 770
},
{
"epoch": 1.95,
"learning_rate": 2.0306406685236768e-05,
"loss": 0.2965,
"step": 780
},
{
"epoch": 1.98,
"learning_rate": 2.0139275766016713e-05,
"loss": 0.3296,
"step": 790
},
{
"epoch": 2.0,
"eval_accuracy": 0.9777875845837011,
"eval_loss": 0.15050138533115387,
"eval_runtime": 13.6335,
"eval_samples_per_second": 498.624,
"eval_steps_per_second": 15.623,
"step": 798
},
{
"epoch": 2.01,
"learning_rate": 1.997214484679666e-05,
"loss": 0.3354,
"step": 800
},
{
"epoch": 2.03,
"learning_rate": 1.9805013927576602e-05,
"loss": 0.3083,
"step": 810
},
{
"epoch": 2.06,
"learning_rate": 1.9637883008356546e-05,
"loss": 0.2819,
"step": 820
},
{
"epoch": 2.08,
"learning_rate": 1.947075208913649e-05,
"loss": 0.2958,
"step": 830
},
{
"epoch": 2.11,
"learning_rate": 1.9303621169916436e-05,
"loss": 0.3153,
"step": 840
},
{
"epoch": 2.13,
"learning_rate": 1.913649025069638e-05,
"loss": 0.2539,
"step": 850
},
{
"epoch": 2.16,
"learning_rate": 1.8969359331476325e-05,
"loss": 0.2776,
"step": 860
},
{
"epoch": 2.18,
"learning_rate": 1.8802228412256266e-05,
"loss": 0.259,
"step": 870
},
{
"epoch": 2.21,
"learning_rate": 1.863509749303621e-05,
"loss": 0.2948,
"step": 880
},
{
"epoch": 2.23,
"learning_rate": 1.846796657381616e-05,
"loss": 0.2331,
"step": 890
},
{
"epoch": 2.26,
"learning_rate": 1.83008356545961e-05,
"loss": 0.2668,
"step": 900
},
{
"epoch": 2.28,
"learning_rate": 1.8133704735376044e-05,
"loss": 0.2456,
"step": 910
},
{
"epoch": 2.31,
"learning_rate": 1.796657381615599e-05,
"loss": 0.2146,
"step": 920
},
{
"epoch": 2.33,
"learning_rate": 1.7799442896935933e-05,
"loss": 0.2564,
"step": 930
},
{
"epoch": 2.36,
"learning_rate": 1.7632311977715878e-05,
"loss": 0.2372,
"step": 940
},
{
"epoch": 2.38,
"learning_rate": 1.7465181058495822e-05,
"loss": 0.2695,
"step": 950
},
{
"epoch": 2.41,
"learning_rate": 1.7298050139275764e-05,
"loss": 0.2716,
"step": 960
},
{
"epoch": 2.43,
"learning_rate": 1.713091922005571e-05,
"loss": 0.244,
"step": 970
},
{
"epoch": 2.46,
"learning_rate": 1.6963788300835656e-05,
"loss": 0.2562,
"step": 980
},
{
"epoch": 2.48,
"learning_rate": 1.6796657381615597e-05,
"loss": 0.2863,
"step": 990
},
{
"epoch": 2.51,
"learning_rate": 1.6629526462395542e-05,
"loss": 0.2448,
"step": 1000
},
{
"epoch": 2.53,
"learning_rate": 1.646239554317549e-05,
"loss": 0.2321,
"step": 1010
},
{
"epoch": 2.56,
"learning_rate": 1.6295264623955434e-05,
"loss": 0.242,
"step": 1020
},
{
"epoch": 2.58,
"learning_rate": 1.6128133704735375e-05,
"loss": 0.272,
"step": 1030
},
{
"epoch": 2.61,
"learning_rate": 1.596100278551532e-05,
"loss": 0.2391,
"step": 1040
},
{
"epoch": 2.63,
"learning_rate": 1.5793871866295268e-05,
"loss": 0.2109,
"step": 1050
},
{
"epoch": 2.66,
"learning_rate": 1.562674094707521e-05,
"loss": 0.2091,
"step": 1060
},
{
"epoch": 2.68,
"learning_rate": 1.5459610027855154e-05,
"loss": 0.2335,
"step": 1070
},
{
"epoch": 2.71,
"learning_rate": 1.5292479108635098e-05,
"loss": 0.1772,
"step": 1080
},
{
"epoch": 2.73,
"learning_rate": 1.5125348189415043e-05,
"loss": 0.2452,
"step": 1090
},
{
"epoch": 2.76,
"learning_rate": 1.4958217270194987e-05,
"loss": 0.2707,
"step": 1100
},
{
"epoch": 2.78,
"learning_rate": 1.479108635097493e-05,
"loss": 0.2312,
"step": 1110
},
{
"epoch": 2.81,
"learning_rate": 1.4623955431754876e-05,
"loss": 0.1965,
"step": 1120
},
{
"epoch": 2.83,
"learning_rate": 1.445682451253482e-05,
"loss": 0.2222,
"step": 1130
},
{
"epoch": 2.86,
"learning_rate": 1.4289693593314764e-05,
"loss": 0.2182,
"step": 1140
},
{
"epoch": 2.88,
"learning_rate": 1.4122562674094708e-05,
"loss": 0.1983,
"step": 1150
},
{
"epoch": 2.91,
"learning_rate": 1.3955431754874653e-05,
"loss": 0.1875,
"step": 1160
},
{
"epoch": 2.93,
"learning_rate": 1.3788300835654596e-05,
"loss": 0.2274,
"step": 1170
},
{
"epoch": 2.96,
"learning_rate": 1.362116991643454e-05,
"loss": 0.205,
"step": 1180
},
{
"epoch": 2.98,
"learning_rate": 1.3454038997214485e-05,
"loss": 0.2088,
"step": 1190
},
{
"epoch": 3.0,
"eval_accuracy": 0.9816122388937923,
"eval_loss": 0.09009595215320587,
"eval_runtime": 13.6443,
"eval_samples_per_second": 498.229,
"eval_steps_per_second": 15.611,
"step": 1197
},
{
"epoch": 3.01,
"learning_rate": 1.3286908077994428e-05,
"loss": 0.2213,
"step": 1200
},
{
"epoch": 3.03,
"learning_rate": 1.3119777158774374e-05,
"loss": 0.231,
"step": 1210
},
{
"epoch": 3.06,
"learning_rate": 1.2952646239554317e-05,
"loss": 0.1848,
"step": 1220
},
{
"epoch": 3.08,
"learning_rate": 1.2785515320334262e-05,
"loss": 0.1874,
"step": 1230
},
{
"epoch": 3.11,
"learning_rate": 1.2618384401114206e-05,
"loss": 0.1669,
"step": 1240
},
{
"epoch": 3.13,
"learning_rate": 1.245125348189415e-05,
"loss": 0.1892,
"step": 1250
},
{
"epoch": 3.16,
"learning_rate": 1.2284122562674095e-05,
"loss": 0.2161,
"step": 1260
},
{
"epoch": 3.18,
"learning_rate": 1.211699164345404e-05,
"loss": 0.2098,
"step": 1270
},
{
"epoch": 3.21,
"learning_rate": 1.1949860724233983e-05,
"loss": 0.1857,
"step": 1280
},
{
"epoch": 3.23,
"learning_rate": 1.1782729805013929e-05,
"loss": 0.2393,
"step": 1290
},
{
"epoch": 3.26,
"learning_rate": 1.1615598885793872e-05,
"loss": 0.2078,
"step": 1300
},
{
"epoch": 3.28,
"learning_rate": 1.1448467966573816e-05,
"loss": 0.2179,
"step": 1310
},
{
"epoch": 3.31,
"learning_rate": 1.1281337047353761e-05,
"loss": 0.1823,
"step": 1320
},
{
"epoch": 3.33,
"learning_rate": 1.1114206128133705e-05,
"loss": 0.2193,
"step": 1330
},
{
"epoch": 3.36,
"learning_rate": 1.0947075208913648e-05,
"loss": 0.1705,
"step": 1340
},
{
"epoch": 3.38,
"learning_rate": 1.0779944289693595e-05,
"loss": 0.2366,
"step": 1350
},
{
"epoch": 3.41,
"learning_rate": 1.0612813370473537e-05,
"loss": 0.1765,
"step": 1360
},
{
"epoch": 3.43,
"learning_rate": 1.0445682451253482e-05,
"loss": 0.1986,
"step": 1370
},
{
"epoch": 3.46,
"learning_rate": 1.0278551532033427e-05,
"loss": 0.257,
"step": 1380
},
{
"epoch": 3.48,
"learning_rate": 1.0111420612813371e-05,
"loss": 0.2109,
"step": 1390
},
{
"epoch": 3.51,
"learning_rate": 9.944289693593314e-06,
"loss": 0.1744,
"step": 1400
},
{
"epoch": 3.53,
"learning_rate": 9.77715877437326e-06,
"loss": 0.1824,
"step": 1410
},
{
"epoch": 3.56,
"learning_rate": 9.610027855153203e-06,
"loss": 0.1797,
"step": 1420
},
{
"epoch": 3.58,
"learning_rate": 9.44289693593315e-06,
"loss": 0.1616,
"step": 1430
},
{
"epoch": 3.61,
"learning_rate": 9.275766016713092e-06,
"loss": 0.1654,
"step": 1440
},
{
"epoch": 3.63,
"learning_rate": 9.108635097493037e-06,
"loss": 0.2256,
"step": 1450
},
{
"epoch": 3.66,
"learning_rate": 8.941504178272981e-06,
"loss": 0.1667,
"step": 1460
},
{
"epoch": 3.68,
"learning_rate": 8.774373259052926e-06,
"loss": 0.1817,
"step": 1470
},
{
"epoch": 3.71,
"learning_rate": 8.607242339832869e-06,
"loss": 0.1801,
"step": 1480
},
{
"epoch": 3.73,
"learning_rate": 8.440111420612815e-06,
"loss": 0.187,
"step": 1490
},
{
"epoch": 3.76,
"learning_rate": 8.272980501392758e-06,
"loss": 0.2072,
"step": 1500
},
{
"epoch": 3.78,
"learning_rate": 8.1058495821727e-06,
"loss": 0.2132,
"step": 1510
},
{
"epoch": 3.81,
"learning_rate": 7.938718662952647e-06,
"loss": 0.165,
"step": 1520
},
{
"epoch": 3.83,
"learning_rate": 7.77158774373259e-06,
"loss": 0.1905,
"step": 1530
},
{
"epoch": 3.86,
"learning_rate": 7.604456824512535e-06,
"loss": 0.223,
"step": 1540
},
{
"epoch": 3.88,
"learning_rate": 7.43732590529248e-06,
"loss": 0.1503,
"step": 1550
},
{
"epoch": 3.91,
"learning_rate": 7.2701949860724235e-06,
"loss": 0.1976,
"step": 1560
},
{
"epoch": 3.93,
"learning_rate": 7.103064066852368e-06,
"loss": 0.1549,
"step": 1570
},
{
"epoch": 3.96,
"learning_rate": 6.935933147632313e-06,
"loss": 0.1742,
"step": 1580
},
{
"epoch": 3.98,
"learning_rate": 6.768802228412256e-06,
"loss": 0.202,
"step": 1590
},
{
"epoch": 4.0,
"eval_accuracy": 0.981318034716093,
"eval_loss": 0.08481906354427338,
"eval_runtime": 13.3788,
"eval_samples_per_second": 508.116,
"eval_steps_per_second": 15.921,
"step": 1596
},
{
"epoch": 4.01,
"learning_rate": 6.601671309192201e-06,
"loss": 0.2132,
"step": 1600
},
{
"epoch": 4.04,
"learning_rate": 6.4345403899721455e-06,
"loss": 0.2056,
"step": 1610
},
{
"epoch": 4.06,
"learning_rate": 6.267409470752089e-06,
"loss": 0.2143,
"step": 1620
},
{
"epoch": 4.09,
"learning_rate": 6.100278551532034e-06,
"loss": 0.166,
"step": 1630
},
{
"epoch": 4.11,
"learning_rate": 5.933147632311978e-06,
"loss": 0.1567,
"step": 1640
},
{
"epoch": 4.14,
"learning_rate": 5.766016713091923e-06,
"loss": 0.1975,
"step": 1650
},
{
"epoch": 4.16,
"learning_rate": 5.598885793871867e-06,
"loss": 0.1609,
"step": 1660
},
{
"epoch": 4.19,
"learning_rate": 5.43175487465181e-06,
"loss": 0.1639,
"step": 1670
},
{
"epoch": 4.21,
"learning_rate": 5.264623955431755e-06,
"loss": 0.1653,
"step": 1680
},
{
"epoch": 4.24,
"learning_rate": 5.0974930362116986e-06,
"loss": 0.1602,
"step": 1690
},
{
"epoch": 4.26,
"learning_rate": 4.930362116991643e-06,
"loss": 0.1981,
"step": 1700
},
{
"epoch": 4.29,
"learning_rate": 4.763231197771588e-06,
"loss": 0.2038,
"step": 1710
},
{
"epoch": 4.31,
"learning_rate": 4.596100278551532e-06,
"loss": 0.1774,
"step": 1720
},
{
"epoch": 4.34,
"learning_rate": 4.428969359331476e-06,
"loss": 0.1846,
"step": 1730
},
{
"epoch": 4.36,
"learning_rate": 4.2618384401114205e-06,
"loss": 0.177,
"step": 1740
},
{
"epoch": 4.39,
"learning_rate": 4.094707520891365e-06,
"loss": 0.2003,
"step": 1750
},
{
"epoch": 4.41,
"learning_rate": 3.927576601671309e-06,
"loss": 0.1663,
"step": 1760
},
{
"epoch": 4.44,
"learning_rate": 3.7604456824512533e-06,
"loss": 0.1597,
"step": 1770
},
{
"epoch": 4.46,
"learning_rate": 3.593314763231198e-06,
"loss": 0.1792,
"step": 1780
},
{
"epoch": 4.49,
"learning_rate": 3.426183844011142e-06,
"loss": 0.1468,
"step": 1790
},
{
"epoch": 4.51,
"learning_rate": 3.259052924791086e-06,
"loss": 0.1833,
"step": 1800
},
{
"epoch": 4.54,
"learning_rate": 3.0919220055710307e-06,
"loss": 0.1762,
"step": 1810
},
{
"epoch": 4.56,
"learning_rate": 2.924791086350975e-06,
"loss": 0.1917,
"step": 1820
},
{
"epoch": 4.59,
"learning_rate": 2.7576601671309194e-06,
"loss": 0.1682,
"step": 1830
},
{
"epoch": 4.61,
"learning_rate": 2.5905292479108636e-06,
"loss": 0.1443,
"step": 1840
},
{
"epoch": 4.64,
"learning_rate": 2.4233983286908077e-06,
"loss": 0.1304,
"step": 1850
},
{
"epoch": 4.66,
"learning_rate": 2.2562674094707523e-06,
"loss": 0.1729,
"step": 1860
},
{
"epoch": 4.69,
"learning_rate": 2.0891364902506964e-06,
"loss": 0.1517,
"step": 1870
},
{
"epoch": 4.71,
"learning_rate": 1.922005571030641e-06,
"loss": 0.1886,
"step": 1880
},
{
"epoch": 4.74,
"learning_rate": 1.7548746518105849e-06,
"loss": 0.1681,
"step": 1890
},
{
"epoch": 4.76,
"learning_rate": 1.5877437325905292e-06,
"loss": 0.1713,
"step": 1900
},
{
"epoch": 4.79,
"learning_rate": 1.4206128133704736e-06,
"loss": 0.2033,
"step": 1910
},
{
"epoch": 4.81,
"learning_rate": 1.253481894150418e-06,
"loss": 0.1564,
"step": 1920
},
{
"epoch": 4.84,
"learning_rate": 1.0863509749303623e-06,
"loss": 0.136,
"step": 1930
},
{
"epoch": 4.86,
"learning_rate": 9.192200557103064e-07,
"loss": 0.1679,
"step": 1940
},
{
"epoch": 4.89,
"learning_rate": 7.520891364902508e-07,
"loss": 0.1564,
"step": 1950
},
{
"epoch": 4.91,
"learning_rate": 5.84958217270195e-07,
"loss": 0.1618,
"step": 1960
},
{
"epoch": 4.94,
"learning_rate": 4.178272980501393e-07,
"loss": 0.164,
"step": 1970
},
{
"epoch": 4.96,
"learning_rate": 2.506963788300836e-07,
"loss": 0.1707,
"step": 1980
},
{
"epoch": 4.99,
"learning_rate": 8.356545961002785e-08,
"loss": 0.1535,
"step": 1990
},
{
"epoch": 5.0,
"eval_accuracy": 0.9819064430714917,
"eval_loss": 0.07737560570240021,
"eval_runtime": 13.6188,
"eval_samples_per_second": 499.163,
"eval_steps_per_second": 15.64,
"step": 1995
},
{
"epoch": 5.0,
"step": 1995,
"total_flos": 2.31918157475328e+18,
"train_loss": 0.546848169783303,
"train_runtime": 857.0555,
"train_samples_per_second": 298.079,
"train_steps_per_second": 2.328
}
],
"max_steps": 1995,
"num_train_epochs": 5,
"total_flos": 2.31918157475328e+18,
"trial_name": null,
"trial_params": null
}