dalio-all-io-1.3b-2-epoch / trainer_state.json
AlekseyKorshuk's picture
End of training
efa59b4
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.0,
"global_step": 58,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 3e-05,
"loss": 2.6543,
"step": 1
},
{
"epoch": 0.03,
"eval_accuracy": 0.05133992343294669,
"eval_loss": 2.611328125,
"eval_runtime": 4.6158,
"eval_samples_per_second": 18.632,
"eval_steps_per_second": 1.3,
"step": 1
},
{
"epoch": 0.07,
"learning_rate": 3e-05,
"loss": 2.6077,
"step": 2
},
{
"epoch": 0.07,
"eval_accuracy": 0.05133992343294669,
"eval_loss": 2.611328125,
"eval_runtime": 4.3958,
"eval_samples_per_second": 19.564,
"eval_steps_per_second": 1.365,
"step": 2
},
{
"epoch": 0.1,
"learning_rate": 2.9978001207766858e-05,
"loss": 2.5964,
"step": 3
},
{
"epoch": 0.1,
"eval_accuracy": 0.05192560425118565,
"eval_loss": 2.560546875,
"eval_runtime": 4.3633,
"eval_samples_per_second": 19.71,
"eval_steps_per_second": 1.375,
"step": 3
},
{
"epoch": 0.14,
"learning_rate": 2.9912069357315394e-05,
"loss": 2.7302,
"step": 4
},
{
"epoch": 0.14,
"eval_accuracy": 0.052668418947488714,
"eval_loss": 2.5234375,
"eval_runtime": 3.6387,
"eval_samples_per_second": 23.635,
"eval_steps_per_second": 1.649,
"step": 4
},
{
"epoch": 0.17,
"learning_rate": 2.9802397838122895e-05,
"loss": 2.7002,
"step": 5
},
{
"epoch": 0.17,
"eval_accuracy": 0.05286840751957031,
"eval_loss": 2.5078125,
"eval_runtime": 4.3812,
"eval_samples_per_second": 19.629,
"eval_steps_per_second": 1.369,
"step": 5
},
{
"epoch": 0.21,
"learning_rate": 2.96493083356513e-05,
"loss": 2.5674,
"step": 6
},
{
"epoch": 0.21,
"eval_accuracy": 0.0532683846637335,
"eval_loss": 2.494140625,
"eval_runtime": 4.3925,
"eval_samples_per_second": 19.579,
"eval_steps_per_second": 1.366,
"step": 6
},
{
"epoch": 0.24,
"learning_rate": 2.9453249887788343e-05,
"loss": 2.6399,
"step": 7
},
{
"epoch": 0.24,
"eval_accuracy": 0.05339694874578595,
"eval_loss": 2.48828125,
"eval_runtime": 4.3565,
"eval_samples_per_second": 19.741,
"eval_steps_per_second": 1.377,
"step": 7
},
{
"epoch": 0.28,
"learning_rate": 2.9214797567742036e-05,
"loss": 2.533,
"step": 8
},
{
"epoch": 0.28,
"eval_accuracy": 0.05362550711387921,
"eval_loss": 2.48046875,
"eval_runtime": 4.3451,
"eval_samples_per_second": 19.792,
"eval_steps_per_second": 1.381,
"step": 8
},
{
"epoch": 0.31,
"learning_rate": 2.8934650797251875e-05,
"loss": 2.7202,
"step": 9
},
{
"epoch": 0.31,
"eval_accuracy": 0.05358265241986172,
"eval_loss": 2.474609375,
"eval_runtime": 4.3538,
"eval_samples_per_second": 19.753,
"eval_steps_per_second": 1.378,
"step": 9
},
{
"epoch": 0.34,
"learning_rate": 2.861363129506436e-05,
"loss": 2.5137,
"step": 10
},
{
"epoch": 0.34,
"eval_accuracy": 0.0533683789497743,
"eval_loss": 2.46484375,
"eval_runtime": 4.2706,
"eval_samples_per_second": 20.138,
"eval_steps_per_second": 1.405,
"step": 10
},
{
"epoch": 0.38,
"learning_rate": 2.8252680666690346e-05,
"loss": 2.499,
"step": 11
},
{
"epoch": 0.38,
"eval_accuracy": 0.053554082623850065,
"eval_loss": 2.451171875,
"eval_runtime": 3.643,
"eval_samples_per_second": 23.607,
"eval_steps_per_second": 1.647,
"step": 11
},
{
"epoch": 0.41,
"learning_rate": 2.7852857642513838e-05,
"loss": 2.7026,
"step": 12
},
{
"epoch": 0.41,
"eval_accuracy": 0.05393977487000743,
"eval_loss": 2.44140625,
"eval_runtime": 4.3592,
"eval_samples_per_second": 19.729,
"eval_steps_per_second": 1.376,
"step": 12
},
{
"epoch": 0.45,
"learning_rate": 2.741533497235336e-05,
"loss": 2.5254,
"step": 13
},
{
"epoch": 0.45,
"eval_accuracy": 0.05431118221815896,
"eval_loss": 2.43359375,
"eval_runtime": 4.3496,
"eval_samples_per_second": 19.772,
"eval_steps_per_second": 1.379,
"step": 13
},
{
"epoch": 0.48,
"learning_rate": 2.6941395985584656e-05,
"loss": 2.5667,
"step": 14
},
{
"epoch": 0.48,
"eval_accuracy": 0.05453974058625222,
"eval_loss": 2.423828125,
"eval_runtime": 3.9428,
"eval_samples_per_second": 21.812,
"eval_steps_per_second": 1.522,
"step": 14
},
{
"epoch": 0.52,
"learning_rate": 2.643243082691455e-05,
"loss": 2.5715,
"step": 15
},
{
"epoch": 0.52,
"eval_accuracy": 0.054768298954345464,
"eval_loss": 2.416015625,
"eval_runtime": 4.3566,
"eval_samples_per_second": 19.74,
"eval_steps_per_second": 1.377,
"step": 15
},
{
"epoch": 0.55,
"learning_rate": 2.5889932378846963e-05,
"loss": 2.3739,
"step": 16
},
{
"epoch": 0.55,
"eval_accuracy": 0.05499685732243872,
"eval_loss": 2.41015625,
"eval_runtime": 3.9638,
"eval_samples_per_second": 21.696,
"eval_steps_per_second": 1.514,
"step": 16
},
{
"epoch": 0.59,
"learning_rate": 2.531549188280135e-05,
"loss": 2.4756,
"step": 17
},
{
"epoch": 0.59,
"eval_accuracy": 0.054896863036397923,
"eval_loss": 2.404296875,
"eval_runtime": 3.6572,
"eval_samples_per_second": 23.515,
"eval_steps_per_second": 1.641,
"step": 17
},
{
"epoch": 0.62,
"learning_rate": 2.4710794271727415e-05,
"loss": 2.4783,
"step": 18
},
{
"epoch": 0.62,
"eval_accuracy": 0.05496828752642706,
"eval_loss": 2.3984375,
"eval_runtime": 4.3549,
"eval_samples_per_second": 19.748,
"eval_steps_per_second": 1.378,
"step": 18
},
{
"epoch": 0.66,
"learning_rate": 2.407761322790648e-05,
"loss": 2.5665,
"step": 19
},
{
"epoch": 0.66,
"eval_accuracy": 0.05488257813839209,
"eval_loss": 2.390625,
"eval_runtime": 4.3461,
"eval_samples_per_second": 19.788,
"eval_steps_per_second": 1.381,
"step": 19
},
{
"epoch": 0.69,
"learning_rate": 2.407761322790648e-05,
"loss": 2.4888,
"step": 20
},
{
"epoch": 0.69,
"eval_accuracy": 0.05488257813839209,
"eval_loss": 2.390625,
"eval_runtime": 4.3835,
"eval_samples_per_second": 19.619,
"eval_steps_per_second": 1.369,
"step": 20
},
{
"epoch": 0.72,
"learning_rate": 2.341780598043574e-05,
"loss": 2.4476,
"step": 21
},
{
"epoch": 0.72,
"eval_accuracy": 0.05499685732243872,
"eval_loss": 2.3828125,
"eval_runtime": 4.2591,
"eval_samples_per_second": 20.192,
"eval_steps_per_second": 1.409,
"step": 21
},
{
"epoch": 0.76,
"learning_rate": 2.2733307857655327e-05,
"loss": 2.604,
"step": 22
},
{
"epoch": 0.76,
"eval_accuracy": 0.05521113079252614,
"eval_loss": 2.375,
"eval_runtime": 4.354,
"eval_samples_per_second": 19.752,
"eval_steps_per_second": 1.378,
"step": 22
},
{
"epoch": 0.79,
"learning_rate": 2.2026126610496852e-05,
"loss": 2.3416,
"step": 23
},
{
"epoch": 0.79,
"eval_accuracy": 0.05539683446660191,
"eval_loss": 2.365234375,
"eval_runtime": 4.3608,
"eval_samples_per_second": 19.721,
"eval_steps_per_second": 1.376,
"step": 23
},
{
"epoch": 0.83,
"learning_rate": 2.129833652340397e-05,
"loss": 2.6028,
"step": 24
},
{
"epoch": 0.83,
"eval_accuracy": 0.05546825895663105,
"eval_loss": 2.35546875,
"eval_runtime": 4.354,
"eval_samples_per_second": 19.752,
"eval_steps_per_second": 1.378,
"step": 24
},
{
"epoch": 0.86,
"learning_rate": 2.055207233009872e-05,
"loss": 2.3425,
"step": 25
},
{
"epoch": 0.86,
"eval_accuracy": 0.055753956916747616,
"eval_loss": 2.34765625,
"eval_runtime": 3.9522,
"eval_samples_per_second": 21.76,
"eval_steps_per_second": 1.518,
"step": 25
},
{
"epoch": 0.9,
"learning_rate": 1.9789522952039697e-05,
"loss": 2.4142,
"step": 26
},
{
"epoch": 0.9,
"eval_accuracy": 0.055811096508770926,
"eval_loss": 2.33984375,
"eval_runtime": 4.3577,
"eval_samples_per_second": 19.735,
"eval_steps_per_second": 1.377,
"step": 26
},
{
"epoch": 0.93,
"learning_rate": 1.9012925077938318e-05,
"loss": 2.5317,
"step": 27
},
{
"epoch": 0.93,
"eval_accuracy": 0.05592537569281755,
"eval_loss": 2.333984375,
"eval_runtime": 3.95,
"eval_samples_per_second": 21.772,
"eval_steps_per_second": 1.519,
"step": 27
},
{
"epoch": 0.97,
"learning_rate": 1.8224556603165363e-05,
"loss": 2.4119,
"step": 28
},
{
"epoch": 0.97,
"eval_accuracy": 0.05612536426489915,
"eval_loss": 2.330078125,
"eval_runtime": 3.959,
"eval_samples_per_second": 21.723,
"eval_steps_per_second": 1.516,
"step": 28
},
{
"epoch": 1.0,
"learning_rate": 1.7426729948291474e-05,
"loss": 2.4048,
"step": 29
},
{
"epoch": 1.0,
"eval_accuracy": 0.0562539283469516,
"eval_loss": 2.326171875,
"eval_runtime": 4.3642,
"eval_samples_per_second": 19.706,
"eval_steps_per_second": 1.375,
"step": 29
},
{
"epoch": 1.03,
"learning_rate": 1.662178527635913e-05,
"loss": 1.9646,
"step": 30
},
{
"epoch": 1.03,
"eval_accuracy": 0.0563539226329924,
"eval_loss": 2.32421875,
"eval_runtime": 4.3686,
"eval_samples_per_second": 19.686,
"eval_steps_per_second": 1.373,
"step": 30
},
{
"epoch": 1.07,
"learning_rate": 1.5812083628781265e-05,
"loss": 1.9233,
"step": 31
},
{
"epoch": 1.07,
"eval_accuracy": 0.056296783040969085,
"eval_loss": 2.3203125,
"eval_runtime": 4.3634,
"eval_samples_per_second": 19.709,
"eval_steps_per_second": 1.375,
"step": 31
},
{
"epoch": 1.1,
"learning_rate": 1.5e-05,
"loss": 1.9276,
"step": 32
},
{
"epoch": 1.1,
"eval_accuracy": 0.0563539226329924,
"eval_loss": 2.3203125,
"eval_runtime": 3.6556,
"eval_samples_per_second": 23.526,
"eval_steps_per_second": 1.641,
"step": 32
},
{
"epoch": 1.14,
"learning_rate": 1.4187916371218739e-05,
"loss": 1.8702,
"step": 33
},
{
"epoch": 1.14,
"eval_accuracy": 0.05649677161305068,
"eval_loss": 2.328125,
"eval_runtime": 3.6562,
"eval_samples_per_second": 23.522,
"eval_steps_per_second": 1.641,
"step": 33
},
{
"epoch": 1.17,
"learning_rate": 1.3378214723640877e-05,
"loss": 2.0997,
"step": 34
},
{
"epoch": 1.17,
"eval_accuracy": 0.05649677161305068,
"eval_loss": 2.333984375,
"eval_runtime": 3.9689,
"eval_samples_per_second": 21.668,
"eval_steps_per_second": 1.512,
"step": 34
},
{
"epoch": 1.21,
"learning_rate": 1.2573270051708529e-05,
"loss": 1.7943,
"step": 35
},
{
"epoch": 1.21,
"eval_accuracy": 0.05676818467516142,
"eval_loss": 2.33203125,
"eval_runtime": 4.3641,
"eval_samples_per_second": 19.706,
"eval_steps_per_second": 1.375,
"step": 35
},
{
"epoch": 1.24,
"learning_rate": 1.177544339683464e-05,
"loss": 1.8579,
"step": 36
},
{
"epoch": 1.24,
"eval_accuracy": 0.056668190389120625,
"eval_loss": 2.32421875,
"eval_runtime": 3.3714,
"eval_samples_per_second": 25.509,
"eval_steps_per_second": 1.78,
"step": 36
},
{
"epoch": 1.28,
"learning_rate": 1.098707492206169e-05,
"loss": 1.8844,
"step": 37
},
{
"epoch": 1.28,
"eval_accuracy": 0.05676818467516142,
"eval_loss": 2.314453125,
"eval_runtime": 4.3578,
"eval_samples_per_second": 19.735,
"eval_steps_per_second": 1.377,
"step": 37
},
{
"epoch": 1.31,
"learning_rate": 1.0210477047960303e-05,
"loss": 1.9288,
"step": 38
},
{
"epoch": 1.31,
"eval_accuracy": 0.05689674875721387,
"eval_loss": 2.30859375,
"eval_runtime": 3.3527,
"eval_samples_per_second": 25.651,
"eval_steps_per_second": 1.79,
"step": 38
},
{
"epoch": 1.34,
"learning_rate": 9.447927669901284e-06,
"loss": 1.6616,
"step": 39
},
{
"epoch": 1.34,
"eval_accuracy": 0.057025312839266325,
"eval_loss": 2.3046875,
"eval_runtime": 4.3576,
"eval_samples_per_second": 19.736,
"eval_steps_per_second": 1.377,
"step": 39
},
{
"epoch": 1.38,
"learning_rate": 8.701663476596031e-06,
"loss": 1.6443,
"step": 40
},
{
"epoch": 1.38,
"eval_accuracy": 0.05705388263527798,
"eval_loss": 2.3046875,
"eval_runtime": 4.3425,
"eval_samples_per_second": 19.804,
"eval_steps_per_second": 1.382,
"step": 40
},
{
"epoch": 1.41,
"learning_rate": 7.97387338950315e-06,
"loss": 1.7616,
"step": 41
},
{
"epoch": 1.41,
"eval_accuracy": 0.057182446717330436,
"eval_loss": 2.302734375,
"eval_runtime": 4.3701,
"eval_samples_per_second": 19.679,
"eval_steps_per_second": 1.373,
"step": 41
},
{
"epoch": 1.45,
"learning_rate": 7.266692142344673e-06,
"loss": 1.7904,
"step": 42
},
{
"epoch": 1.45,
"eval_accuracy": 0.0571110222273013,
"eval_loss": 2.302734375,
"eval_runtime": 3.9729,
"eval_samples_per_second": 21.647,
"eval_steps_per_second": 1.51,
"step": 42
},
{
"epoch": 1.48,
"learning_rate": 6.582194019564266e-06,
"loss": 1.8762,
"step": 43
},
{
"epoch": 1.48,
"eval_accuracy": 0.05729672590137706,
"eval_loss": 2.302734375,
"eval_runtime": 4.3645,
"eval_samples_per_second": 19.704,
"eval_steps_per_second": 1.375,
"step": 43
},
{
"epoch": 1.52,
"learning_rate": 5.922386772093526e-06,
"loss": 1.6569,
"step": 44
},
{
"epoch": 1.52,
"eval_accuracy": 0.05725387120735958,
"eval_loss": 2.302734375,
"eval_runtime": 4.4596,
"eval_samples_per_second": 19.284,
"eval_steps_per_second": 1.345,
"step": 44
},
{
"epoch": 1.55,
"learning_rate": 5.289205728272587e-06,
"loss": 1.647,
"step": 45
},
{
"epoch": 1.55,
"eval_accuracy": 0.05733958059539455,
"eval_loss": 2.302734375,
"eval_runtime": 4.3534,
"eval_samples_per_second": 19.755,
"eval_steps_per_second": 1.378,
"step": 45
},
{
"epoch": 1.59,
"learning_rate": 4.684508117198649e-06,
"loss": 1.8168,
"step": 46
},
{
"epoch": 1.59,
"eval_accuracy": 0.057425289983429516,
"eval_loss": 2.302734375,
"eval_runtime": 3.9668,
"eval_samples_per_second": 21.68,
"eval_steps_per_second": 1.513,
"step": 46
},
{
"epoch": 1.62,
"learning_rate": 4.110067621153041e-06,
"loss": 1.7194,
"step": 47
},
{
"epoch": 1.62,
"eval_accuracy": 0.057311010799382896,
"eval_loss": 2.302734375,
"eval_runtime": 4.3648,
"eval_samples_per_second": 19.703,
"eval_steps_per_second": 1.375,
"step": 47
},
{
"epoch": 1.66,
"learning_rate": 3.567569173085455e-06,
"loss": 1.7667,
"step": 48
},
{
"epoch": 1.66,
"eval_accuracy": 0.05722530141134792,
"eval_loss": 2.302734375,
"eval_runtime": 4.3629,
"eval_samples_per_second": 19.712,
"eval_steps_per_second": 1.375,
"step": 48
},
{
"epoch": 1.69,
"learning_rate": 3.0586040144153436e-06,
"loss": 1.7621,
"step": 49
},
{
"epoch": 1.69,
"eval_accuracy": 0.05729672590137706,
"eval_loss": 2.302734375,
"eval_runtime": 4.3603,
"eval_samples_per_second": 19.723,
"eval_steps_per_second": 1.376,
"step": 49
},
{
"epoch": 1.72,
"learning_rate": 2.5846650276466435e-06,
"loss": 1.7269,
"step": 50
},
{
"epoch": 1.72,
"eval_accuracy": 0.05733958059539455,
"eval_loss": 2.30078125,
"eval_runtime": 3.6679,
"eval_samples_per_second": 23.447,
"eval_steps_per_second": 1.636,
"step": 50
},
{
"epoch": 1.76,
"learning_rate": 2.1471423574861643e-06,
"loss": 1.7815,
"step": 51
},
{
"epoch": 1.76,
"eval_accuracy": 0.057396720187417864,
"eval_loss": 2.30078125,
"eval_runtime": 4.3528,
"eval_samples_per_second": 19.757,
"eval_steps_per_second": 1.378,
"step": 51
},
{
"epoch": 1.79,
"learning_rate": 1.7473193333096576e-06,
"loss": 1.8318,
"step": 52
},
{
"epoch": 1.79,
"eval_accuracy": 0.057396720187417864,
"eval_loss": 2.298828125,
"eval_runtime": 4.3561,
"eval_samples_per_second": 19.743,
"eval_steps_per_second": 1.377,
"step": 52
},
{
"epoch": 1.83,
"learning_rate": 1.3863687049356465e-06,
"loss": 1.9366,
"step": 53
},
{
"epoch": 1.83,
"eval_accuracy": 0.057468144677447,
"eval_loss": 2.298828125,
"eval_runtime": 4.3614,
"eval_samples_per_second": 19.719,
"eval_steps_per_second": 1.376,
"step": 53
},
{
"epoch": 1.86,
"learning_rate": 1.0653492027481287e-06,
"loss": 1.736,
"step": 54
},
{
"epoch": 1.86,
"eval_accuracy": 0.057553854065481976,
"eval_loss": 2.296875,
"eval_runtime": 4.3622,
"eval_samples_per_second": 19.715,
"eval_steps_per_second": 1.375,
"step": 54
},
{
"epoch": 1.9,
"learning_rate": 7.852024322579649e-07,
"loss": 1.9984,
"step": 55
},
{
"epoch": 1.9,
"eval_accuracy": 0.05753956916747614,
"eval_loss": 2.296875,
"eval_runtime": 4.3417,
"eval_samples_per_second": 19.808,
"eval_steps_per_second": 1.382,
"step": 55
},
{
"epoch": 1.93,
"learning_rate": 5.467501122116564e-07,
"loss": 1.7203,
"step": 56
},
{
"epoch": 1.93,
"eval_accuracy": 0.05753956916747614,
"eval_loss": 2.294921875,
"eval_runtime": 3.6633,
"eval_samples_per_second": 23.476,
"eval_steps_per_second": 1.638,
"step": 56
},
{
"epoch": 1.97,
"learning_rate": 3.5069166434870014e-07,
"loss": 1.7391,
"step": 57
},
{
"epoch": 1.97,
"eval_accuracy": 0.0575681389634878,
"eval_loss": 2.294921875,
"eval_runtime": 4.3729,
"eval_samples_per_second": 19.667,
"eval_steps_per_second": 1.372,
"step": 57
},
{
"epoch": 2.0,
"learning_rate": 1.9760216187710788e-07,
"loss": 1.6611,
"step": 58
},
{
"epoch": 2.0,
"eval_accuracy": 0.057553854065481976,
"eval_loss": 2.294921875,
"eval_runtime": 3.3519,
"eval_samples_per_second": 25.657,
"eval_steps_per_second": 1.79,
"step": 58
},
{
"epoch": 2.0,
"step": 58,
"total_flos": 3170519875584.0,
"train_loss": 2.1741048878636855,
"train_runtime": 774.6069,
"train_samples_per_second": 1.185,
"train_steps_per_second": 0.075
}
],
"max_steps": 58,
"num_train_epochs": 2,
"total_flos": 3170519875584.0,
"trial_name": null,
"trial_params": null
}