phi-1_5_sft / trainer_state.json
hushell's picture
Model save
d9da186 verified
{
"best_metric": 1.254175066947937,
"best_model_checkpoint": "data/phi-1_5_sft/checkpoint-2900",
"epoch": 2.998499249624812,
"eval_steps": 100,
"global_step": 2997,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 1.6666666666666668e-07,
"loss": 1.6512,
"step": 1
},
{
"epoch": 0.01,
"learning_rate": 8.333333333333333e-07,
"loss": 1.7421,
"step": 5
},
{
"epoch": 0.01,
"learning_rate": 1.6666666666666667e-06,
"loss": 1.7114,
"step": 10
},
{
"epoch": 0.02,
"learning_rate": 2.5e-06,
"loss": 1.6882,
"step": 15
},
{
"epoch": 0.02,
"learning_rate": 3.3333333333333333e-06,
"loss": 1.5959,
"step": 20
},
{
"epoch": 0.03,
"learning_rate": 4.166666666666667e-06,
"loss": 1.4829,
"step": 25
},
{
"epoch": 0.03,
"learning_rate": 5e-06,
"loss": 1.4395,
"step": 30
},
{
"epoch": 0.04,
"learning_rate": 5.833333333333334e-06,
"loss": 1.4149,
"step": 35
},
{
"epoch": 0.04,
"learning_rate": 6.666666666666667e-06,
"loss": 1.3751,
"step": 40
},
{
"epoch": 0.05,
"learning_rate": 7.500000000000001e-06,
"loss": 1.3746,
"step": 45
},
{
"epoch": 0.05,
"learning_rate": 8.333333333333334e-06,
"loss": 1.3864,
"step": 50
},
{
"epoch": 0.06,
"learning_rate": 9.166666666666666e-06,
"loss": 1.3487,
"step": 55
},
{
"epoch": 0.06,
"learning_rate": 1e-05,
"loss": 1.3613,
"step": 60
},
{
"epoch": 0.07,
"learning_rate": 1.0833333333333334e-05,
"loss": 1.3476,
"step": 65
},
{
"epoch": 0.07,
"learning_rate": 1.1666666666666668e-05,
"loss": 1.3487,
"step": 70
},
{
"epoch": 0.08,
"learning_rate": 1.25e-05,
"loss": 1.3485,
"step": 75
},
{
"epoch": 0.08,
"learning_rate": 1.3333333333333333e-05,
"loss": 1.347,
"step": 80
},
{
"epoch": 0.09,
"learning_rate": 1.416666666666667e-05,
"loss": 1.3166,
"step": 85
},
{
"epoch": 0.09,
"learning_rate": 1.5000000000000002e-05,
"loss": 1.3199,
"step": 90
},
{
"epoch": 0.1,
"learning_rate": 1.5833333333333333e-05,
"loss": 1.3265,
"step": 95
},
{
"epoch": 0.1,
"learning_rate": 1.6666666666666667e-05,
"loss": 1.3099,
"step": 100
},
{
"epoch": 0.1,
"eval_loss": 1.339752197265625,
"eval_runtime": 152.8603,
"eval_samples_per_second": 92.581,
"eval_steps_per_second": 2.898,
"step": 100
},
{
"epoch": 0.11,
"learning_rate": 1.7500000000000002e-05,
"loss": 1.3124,
"step": 105
},
{
"epoch": 0.11,
"learning_rate": 1.8333333333333333e-05,
"loss": 1.3422,
"step": 110
},
{
"epoch": 0.12,
"learning_rate": 1.916666666666667e-05,
"loss": 1.3155,
"step": 115
},
{
"epoch": 0.12,
"learning_rate": 2e-05,
"loss": 1.3181,
"step": 120
},
{
"epoch": 0.13,
"learning_rate": 1.9999850951038044e-05,
"loss": 1.296,
"step": 125
},
{
"epoch": 0.13,
"learning_rate": 1.999940380859529e-05,
"loss": 1.2947,
"step": 130
},
{
"epoch": 0.14,
"learning_rate": 1.9998658586000968e-05,
"loss": 1.3271,
"step": 135
},
{
"epoch": 0.14,
"learning_rate": 1.9997615305470002e-05,
"loss": 1.3193,
"step": 140
},
{
"epoch": 0.15,
"learning_rate": 1.9996273998102368e-05,
"loss": 1.321,
"step": 145
},
{
"epoch": 0.15,
"learning_rate": 1.9994634703882166e-05,
"loss": 1.3308,
"step": 150
},
{
"epoch": 0.16,
"learning_rate": 1.9992697471676413e-05,
"loss": 1.329,
"step": 155
},
{
"epoch": 0.16,
"learning_rate": 1.9990462359233595e-05,
"loss": 1.324,
"step": 160
},
{
"epoch": 0.17,
"learning_rate": 1.998792943318196e-05,
"loss": 1.3263,
"step": 165
},
{
"epoch": 0.17,
"learning_rate": 1.99850987690275e-05,
"loss": 1.3239,
"step": 170
},
{
"epoch": 0.18,
"learning_rate": 1.9981970451151724e-05,
"loss": 1.3088,
"step": 175
},
{
"epoch": 0.18,
"learning_rate": 1.9978544572809143e-05,
"loss": 1.3385,
"step": 180
},
{
"epoch": 0.19,
"learning_rate": 1.997482123612448e-05,
"loss": 1.3323,
"step": 185
},
{
"epoch": 0.19,
"learning_rate": 1.9970800552089623e-05,
"loss": 1.308,
"step": 190
},
{
"epoch": 0.2,
"learning_rate": 1.9966482640560335e-05,
"loss": 1.2985,
"step": 195
},
{
"epoch": 0.2,
"learning_rate": 1.996186763025266e-05,
"loss": 1.3131,
"step": 200
},
{
"epoch": 0.2,
"eval_loss": 1.315926194190979,
"eval_runtime": 152.9359,
"eval_samples_per_second": 92.535,
"eval_steps_per_second": 2.897,
"step": 200
},
{
"epoch": 0.21,
"learning_rate": 1.9956955658739098e-05,
"loss": 1.3049,
"step": 205
},
{
"epoch": 0.21,
"learning_rate": 1.995174687244449e-05,
"loss": 1.3123,
"step": 210
},
{
"epoch": 0.22,
"learning_rate": 1.994624142664169e-05,
"loss": 1.3071,
"step": 215
},
{
"epoch": 0.22,
"learning_rate": 1.9940439485446884e-05,
"loss": 1.2991,
"step": 220
},
{
"epoch": 0.23,
"learning_rate": 1.993434122181474e-05,
"loss": 1.3026,
"step": 225
},
{
"epoch": 0.23,
"learning_rate": 1.9927946817533224e-05,
"loss": 1.2919,
"step": 230
},
{
"epoch": 0.24,
"learning_rate": 1.9921256463218207e-05,
"loss": 1.2949,
"step": 235
},
{
"epoch": 0.24,
"learning_rate": 1.9914270358307762e-05,
"loss": 1.3001,
"step": 240
},
{
"epoch": 0.25,
"learning_rate": 1.9906988711056215e-05,
"loss": 1.3064,
"step": 245
},
{
"epoch": 0.25,
"learning_rate": 1.9899411738527977e-05,
"loss": 1.3145,
"step": 250
},
{
"epoch": 0.26,
"learning_rate": 1.989153966659101e-05,
"loss": 1.3119,
"step": 255
},
{
"epoch": 0.26,
"learning_rate": 1.9883372729910154e-05,
"loss": 1.2871,
"step": 260
},
{
"epoch": 0.27,
"learning_rate": 1.9874911171940087e-05,
"loss": 1.2849,
"step": 265
},
{
"epoch": 0.27,
"learning_rate": 1.9866155244918106e-05,
"loss": 1.3173,
"step": 270
},
{
"epoch": 0.28,
"learning_rate": 1.985710520985657e-05,
"loss": 1.285,
"step": 275
},
{
"epoch": 0.28,
"learning_rate": 1.9847761336535148e-05,
"loss": 1.2926,
"step": 280
},
{
"epoch": 0.29,
"learning_rate": 1.9838123903492764e-05,
"loss": 1.2915,
"step": 285
},
{
"epoch": 0.29,
"learning_rate": 1.9828193198019297e-05,
"loss": 1.2786,
"step": 290
},
{
"epoch": 0.3,
"learning_rate": 1.981796951614701e-05,
"loss": 1.2994,
"step": 295
},
{
"epoch": 0.3,
"learning_rate": 1.9807453162641745e-05,
"loss": 1.3009,
"step": 300
},
{
"epoch": 0.3,
"eval_loss": 1.3045789003372192,
"eval_runtime": 152.7904,
"eval_samples_per_second": 92.624,
"eval_steps_per_second": 2.899,
"step": 300
},
{
"epoch": 0.31,
"learning_rate": 1.9796644450993812e-05,
"loss": 1.274,
"step": 305
},
{
"epoch": 0.31,
"learning_rate": 1.978554370340866e-05,
"loss": 1.2899,
"step": 310
},
{
"epoch": 0.32,
"learning_rate": 1.9774151250797278e-05,
"loss": 1.311,
"step": 315
},
{
"epoch": 0.32,
"learning_rate": 1.9762467432766303e-05,
"loss": 1.2917,
"step": 320
},
{
"epoch": 0.33,
"learning_rate": 1.975049259760793e-05,
"loss": 1.2636,
"step": 325
},
{
"epoch": 0.33,
"learning_rate": 1.973822710228951e-05,
"loss": 1.305,
"step": 330
},
{
"epoch": 0.34,
"learning_rate": 1.972567131244291e-05,
"loss": 1.2816,
"step": 335
},
{
"epoch": 0.34,
"learning_rate": 1.9712825602353617e-05,
"loss": 1.2708,
"step": 340
},
{
"epoch": 0.35,
"learning_rate": 1.969969035494958e-05,
"loss": 1.2657,
"step": 345
},
{
"epoch": 0.35,
"learning_rate": 1.9686265961789807e-05,
"loss": 1.2958,
"step": 350
},
{
"epoch": 0.36,
"learning_rate": 1.9672552823052663e-05,
"loss": 1.2901,
"step": 355
},
{
"epoch": 0.36,
"learning_rate": 1.965855134752397e-05,
"loss": 1.2807,
"step": 360
},
{
"epoch": 0.37,
"learning_rate": 1.9644261952584804e-05,
"loss": 1.2744,
"step": 365
},
{
"epoch": 0.37,
"learning_rate": 1.962968506419906e-05,
"loss": 1.2891,
"step": 370
},
{
"epoch": 0.38,
"learning_rate": 1.9614821116900763e-05,
"loss": 1.2947,
"step": 375
},
{
"epoch": 0.38,
"learning_rate": 1.9599670553781087e-05,
"loss": 1.251,
"step": 380
},
{
"epoch": 0.39,
"learning_rate": 1.9584233826475178e-05,
"loss": 1.2793,
"step": 385
},
{
"epoch": 0.39,
"learning_rate": 1.9568511395148674e-05,
"loss": 1.2735,
"step": 390
},
{
"epoch": 0.4,
"learning_rate": 1.955250372848398e-05,
"loss": 1.2796,
"step": 395
},
{
"epoch": 0.4,
"learning_rate": 1.9536211303666324e-05,
"loss": 1.2915,
"step": 400
},
{
"epoch": 0.4,
"eval_loss": 1.296703577041626,
"eval_runtime": 152.879,
"eval_samples_per_second": 92.57,
"eval_steps_per_second": 2.898,
"step": 400
},
{
"epoch": 0.41,
"learning_rate": 1.9519634606369506e-05,
"loss": 1.2607,
"step": 405
},
{
"epoch": 0.41,
"learning_rate": 1.9502774130741427e-05,
"loss": 1.3004,
"step": 410
},
{
"epoch": 0.42,
"learning_rate": 1.948563037938937e-05,
"loss": 1.2881,
"step": 415
},
{
"epoch": 0.42,
"learning_rate": 1.9468203863365002e-05,
"loss": 1.2765,
"step": 420
},
{
"epoch": 0.43,
"learning_rate": 1.9450495102149144e-05,
"loss": 1.2955,
"step": 425
},
{
"epoch": 0.43,
"learning_rate": 1.94325046236363e-05,
"loss": 1.3082,
"step": 430
},
{
"epoch": 0.44,
"learning_rate": 1.9414232964118893e-05,
"loss": 1.2897,
"step": 435
},
{
"epoch": 0.44,
"learning_rate": 1.93956806682713e-05,
"loss": 1.287,
"step": 440
},
{
"epoch": 0.45,
"learning_rate": 1.9376848289133607e-05,
"loss": 1.2774,
"step": 445
},
{
"epoch": 0.45,
"learning_rate": 1.9357736388095132e-05,
"loss": 1.2961,
"step": 450
},
{
"epoch": 0.46,
"learning_rate": 1.9338345534877675e-05,
"loss": 1.2771,
"step": 455
},
{
"epoch": 0.46,
"learning_rate": 1.9318676307518543e-05,
"loss": 1.2679,
"step": 460
},
{
"epoch": 0.47,
"learning_rate": 1.9298729292353318e-05,
"loss": 1.3019,
"step": 465
},
{
"epoch": 0.47,
"learning_rate": 1.9278505083998393e-05,
"loss": 1.2785,
"step": 470
},
{
"epoch": 0.48,
"learning_rate": 1.9258004285333204e-05,
"loss": 1.2741,
"step": 475
},
{
"epoch": 0.48,
"learning_rate": 1.923722750748231e-05,
"loss": 1.281,
"step": 480
},
{
"epoch": 0.49,
"learning_rate": 1.9216175369797145e-05,
"loss": 1.2836,
"step": 485
},
{
"epoch": 0.49,
"learning_rate": 1.9194848499837567e-05,
"loss": 1.292,
"step": 490
},
{
"epoch": 0.5,
"learning_rate": 1.9173247533353137e-05,
"loss": 1.2937,
"step": 495
},
{
"epoch": 0.5,
"learning_rate": 1.9151373114264183e-05,
"loss": 1.2714,
"step": 500
},
{
"epoch": 0.5,
"eval_loss": 1.2905510663986206,
"eval_runtime": 152.8556,
"eval_samples_per_second": 92.584,
"eval_steps_per_second": 2.898,
"step": 500
},
{
"epoch": 0.51,
"learning_rate": 1.9129225894642594e-05,
"loss": 1.2424,
"step": 505
},
{
"epoch": 0.51,
"learning_rate": 1.9106806534692392e-05,
"loss": 1.2737,
"step": 510
},
{
"epoch": 0.52,
"learning_rate": 1.9084115702730043e-05,
"loss": 1.268,
"step": 515
},
{
"epoch": 0.52,
"learning_rate": 1.9061154075164535e-05,
"loss": 1.2651,
"step": 520
},
{
"epoch": 0.53,
"learning_rate": 1.903792233647722e-05,
"loss": 1.2828,
"step": 525
},
{
"epoch": 0.53,
"learning_rate": 1.9014421179201407e-05,
"loss": 1.2656,
"step": 530
},
{
"epoch": 0.54,
"learning_rate": 1.899065130390171e-05,
"loss": 1.2667,
"step": 535
},
{
"epoch": 0.54,
"learning_rate": 1.896661341915318e-05,
"loss": 1.3043,
"step": 540
},
{
"epoch": 0.55,
"learning_rate": 1.8942308241520174e-05,
"loss": 1.2754,
"step": 545
},
{
"epoch": 0.55,
"learning_rate": 1.8917736495534988e-05,
"loss": 1.2773,
"step": 550
},
{
"epoch": 0.56,
"learning_rate": 1.8892898913676264e-05,
"loss": 1.2737,
"step": 555
},
{
"epoch": 0.56,
"learning_rate": 1.8867796236347168e-05,
"loss": 1.2881,
"step": 560
},
{
"epoch": 0.57,
"learning_rate": 1.8842429211853298e-05,
"loss": 1.2547,
"step": 565
},
{
"epoch": 0.57,
"learning_rate": 1.8816798596380388e-05,
"loss": 1.256,
"step": 570
},
{
"epoch": 0.58,
"learning_rate": 1.879090515397176e-05,
"loss": 1.291,
"step": 575
},
{
"epoch": 0.58,
"learning_rate": 1.876474965650556e-05,
"loss": 1.2748,
"step": 580
},
{
"epoch": 0.59,
"learning_rate": 1.8738332883671738e-05,
"loss": 1.2653,
"step": 585
},
{
"epoch": 0.59,
"learning_rate": 1.8711655622948805e-05,
"loss": 1.2727,
"step": 590
},
{
"epoch": 0.6,
"learning_rate": 1.8684718669580368e-05,
"loss": 1.2425,
"step": 595
},
{
"epoch": 0.6,
"learning_rate": 1.8657522826551412e-05,
"loss": 1.2811,
"step": 600
},
{
"epoch": 0.6,
"eval_loss": 1.2853693962097168,
"eval_runtime": 152.8403,
"eval_samples_per_second": 92.593,
"eval_steps_per_second": 2.898,
"step": 600
},
{
"epoch": 0.61,
"learning_rate": 1.8630068904564373e-05,
"loss": 1.2578,
"step": 605
},
{
"epoch": 0.61,
"learning_rate": 1.8602357722014966e-05,
"loss": 1.254,
"step": 610
},
{
"epoch": 0.62,
"learning_rate": 1.8574390104967784e-05,
"loss": 1.2757,
"step": 615
},
{
"epoch": 0.62,
"learning_rate": 1.85461668871317e-05,
"loss": 1.2756,
"step": 620
},
{
"epoch": 0.63,
"learning_rate": 1.851768890983496e-05,
"loss": 1.2673,
"step": 625
},
{
"epoch": 0.63,
"learning_rate": 1.8488957022000164e-05,
"loss": 1.2752,
"step": 630
},
{
"epoch": 0.64,
"learning_rate": 1.8459972080118925e-05,
"loss": 1.2691,
"step": 635
},
{
"epoch": 0.64,
"learning_rate": 1.8430734948226336e-05,
"loss": 1.2737,
"step": 640
},
{
"epoch": 0.65,
"learning_rate": 1.8401246497875238e-05,
"loss": 1.2601,
"step": 645
},
{
"epoch": 0.65,
"learning_rate": 1.8371507608110206e-05,
"loss": 1.2711,
"step": 650
},
{
"epoch": 0.66,
"learning_rate": 1.8341519165441373e-05,
"loss": 1.2861,
"step": 655
},
{
"epoch": 0.66,
"learning_rate": 1.831128206381799e-05,
"loss": 1.2867,
"step": 660
},
{
"epoch": 0.67,
"learning_rate": 1.8280797204601782e-05,
"loss": 1.2666,
"step": 665
},
{
"epoch": 0.67,
"learning_rate": 1.8250065496540062e-05,
"loss": 1.2876,
"step": 670
},
{
"epoch": 0.68,
"learning_rate": 1.8219087855738678e-05,
"loss": 1.2693,
"step": 675
},
{
"epoch": 0.68,
"learning_rate": 1.818786520563467e-05,
"loss": 1.2868,
"step": 680
},
{
"epoch": 0.69,
"learning_rate": 1.815639847696875e-05,
"loss": 1.252,
"step": 685
},
{
"epoch": 0.69,
"learning_rate": 1.812468860775757e-05,
"loss": 1.2457,
"step": 690
},
{
"epoch": 0.7,
"learning_rate": 1.809273654326575e-05,
"loss": 1.2627,
"step": 695
},
{
"epoch": 0.7,
"learning_rate": 1.8060543235977696e-05,
"loss": 1.2621,
"step": 700
},
{
"epoch": 0.7,
"eval_loss": 1.2807434797286987,
"eval_runtime": 152.8742,
"eval_samples_per_second": 92.573,
"eval_steps_per_second": 2.898,
"step": 700
},
{
"epoch": 0.71,
"learning_rate": 1.8028109645569212e-05,
"loss": 1.2517,
"step": 705
},
{
"epoch": 0.71,
"learning_rate": 1.79954367388789e-05,
"loss": 1.2717,
"step": 710
},
{
"epoch": 0.72,
"learning_rate": 1.7962525489879324e-05,
"loss": 1.2707,
"step": 715
},
{
"epoch": 0.72,
"learning_rate": 1.7929376879647984e-05,
"loss": 1.2745,
"step": 720
},
{
"epoch": 0.73,
"learning_rate": 1.7895991896338066e-05,
"loss": 1.2715,
"step": 725
},
{
"epoch": 0.73,
"learning_rate": 1.7862371535149e-05,
"loss": 1.2419,
"step": 730
},
{
"epoch": 0.74,
"learning_rate": 1.782851679829676e-05,
"loss": 1.2804,
"step": 735
},
{
"epoch": 0.74,
"learning_rate": 1.779442869498403e-05,
"loss": 1.2821,
"step": 740
},
{
"epoch": 0.75,
"learning_rate": 1.7760108241370097e-05,
"loss": 1.2569,
"step": 745
},
{
"epoch": 0.75,
"learning_rate": 1.7725556460540553e-05,
"loss": 1.2802,
"step": 750
},
{
"epoch": 0.76,
"learning_rate": 1.7690774382476808e-05,
"loss": 1.258,
"step": 755
},
{
"epoch": 0.76,
"learning_rate": 1.7655763044025395e-05,
"loss": 1.2775,
"step": 760
},
{
"epoch": 0.77,
"learning_rate": 1.7620523488867042e-05,
"loss": 1.2684,
"step": 765
},
{
"epoch": 0.77,
"learning_rate": 1.7585056767485574e-05,
"loss": 1.2835,
"step": 770
},
{
"epoch": 0.78,
"learning_rate": 1.7549363937136586e-05,
"loss": 1.246,
"step": 775
},
{
"epoch": 0.78,
"learning_rate": 1.7513446061815946e-05,
"loss": 1.2623,
"step": 780
},
{
"epoch": 0.79,
"learning_rate": 1.747730421222806e-05,
"loss": 1.2679,
"step": 785
},
{
"epoch": 0.79,
"learning_rate": 1.744093946575396e-05,
"loss": 1.2835,
"step": 790
},
{
"epoch": 0.8,
"learning_rate": 1.7404352906419187e-05,
"loss": 1.2549,
"step": 795
},
{
"epoch": 0.8,
"learning_rate": 1.7367545624861483e-05,
"loss": 1.2406,
"step": 800
},
{
"epoch": 0.8,
"eval_loss": 1.2767128944396973,
"eval_runtime": 152.8153,
"eval_samples_per_second": 92.609,
"eval_steps_per_second": 2.899,
"step": 800
},
{
"epoch": 0.81,
"learning_rate": 1.7330518718298263e-05,
"loss": 1.2444,
"step": 805
},
{
"epoch": 0.81,
"learning_rate": 1.7293273290493932e-05,
"loss": 1.2615,
"step": 810
},
{
"epoch": 0.82,
"learning_rate": 1.725581045172696e-05,
"loss": 1.2722,
"step": 815
},
{
"epoch": 0.82,
"learning_rate": 1.721813131875679e-05,
"loss": 1.2676,
"step": 820
},
{
"epoch": 0.83,
"learning_rate": 1.718023701479055e-05,
"loss": 1.2455,
"step": 825
},
{
"epoch": 0.83,
"learning_rate": 1.714212866944958e-05,
"loss": 1.2792,
"step": 830
},
{
"epoch": 0.84,
"learning_rate": 1.710380741873574e-05,
"loss": 1.2674,
"step": 835
},
{
"epoch": 0.84,
"learning_rate": 1.706527440499756e-05,
"loss": 1.2514,
"step": 840
},
{
"epoch": 0.85,
"learning_rate": 1.702653077689618e-05,
"loss": 1.2564,
"step": 845
},
{
"epoch": 0.85,
"learning_rate": 1.6987577689371107e-05,
"loss": 1.2636,
"step": 850
},
{
"epoch": 0.86,
"learning_rate": 1.6948416303605796e-05,
"loss": 1.2583,
"step": 855
},
{
"epoch": 0.86,
"learning_rate": 1.690904778699302e-05,
"loss": 1.2724,
"step": 860
},
{
"epoch": 0.87,
"learning_rate": 1.68694733131001e-05,
"loss": 1.2631,
"step": 865
},
{
"epoch": 0.87,
"learning_rate": 1.6829694061633874e-05,
"loss": 1.2713,
"step": 870
},
{
"epoch": 0.88,
"learning_rate": 1.6789711218405575e-05,
"loss": 1.2453,
"step": 875
},
{
"epoch": 0.88,
"learning_rate": 1.6749525975295455e-05,
"loss": 1.2941,
"step": 880
},
{
"epoch": 0.89,
"learning_rate": 1.670913953021727e-05,
"loss": 1.2435,
"step": 885
},
{
"epoch": 0.89,
"learning_rate": 1.666855308708257e-05,
"loss": 1.261,
"step": 890
},
{
"epoch": 0.9,
"learning_rate": 1.6627767855764786e-05,
"loss": 1.2629,
"step": 895
},
{
"epoch": 0.9,
"learning_rate": 1.6586785052063206e-05,
"loss": 1.2371,
"step": 900
},
{
"epoch": 0.9,
"eval_loss": 1.273120403289795,
"eval_runtime": 152.8358,
"eval_samples_per_second": 92.596,
"eval_steps_per_second": 2.899,
"step": 900
},
{
"epoch": 0.91,
"learning_rate": 1.65456058976667e-05,
"loss": 1.2868,
"step": 905
},
{
"epoch": 0.91,
"learning_rate": 1.6504231620117305e-05,
"loss": 1.2628,
"step": 910
},
{
"epoch": 0.92,
"learning_rate": 1.646266345277365e-05,
"loss": 1.2489,
"step": 915
},
{
"epoch": 0.92,
"learning_rate": 1.642090263477417e-05,
"loss": 1.2645,
"step": 920
},
{
"epoch": 0.93,
"learning_rate": 1.6378950411000183e-05,
"loss": 1.2803,
"step": 925
},
{
"epoch": 0.93,
"learning_rate": 1.6336808032038768e-05,
"loss": 1.2429,
"step": 930
},
{
"epoch": 0.94,
"learning_rate": 1.6294476754145497e-05,
"loss": 1.2285,
"step": 935
},
{
"epoch": 0.94,
"learning_rate": 1.625195783920697e-05,
"loss": 1.2508,
"step": 940
},
{
"epoch": 0.95,
"learning_rate": 1.6209252554703215e-05,
"loss": 1.2661,
"step": 945
},
{
"epoch": 0.95,
"learning_rate": 1.61663621736699e-05,
"loss": 1.2709,
"step": 950
},
{
"epoch": 0.96,
"learning_rate": 1.6123287974660378e-05,
"loss": 1.25,
"step": 955
},
{
"epoch": 0.96,
"learning_rate": 1.608003124170758e-05,
"loss": 1.2693,
"step": 960
},
{
"epoch": 0.97,
"learning_rate": 1.603659326428573e-05,
"loss": 1.2574,
"step": 965
},
{
"epoch": 0.97,
"learning_rate": 1.599297533727192e-05,
"loss": 1.2554,
"step": 970
},
{
"epoch": 0.98,
"learning_rate": 1.59491787609075e-05,
"loss": 1.2678,
"step": 975
},
{
"epoch": 0.98,
"learning_rate": 1.590520484075932e-05,
"loss": 1.242,
"step": 980
},
{
"epoch": 0.99,
"learning_rate": 1.5861054887680805e-05,
"loss": 1.2637,
"step": 985
},
{
"epoch": 0.99,
"learning_rate": 1.581673021777289e-05,
"loss": 1.2353,
"step": 990
},
{
"epoch": 1.0,
"learning_rate": 1.5772232152344797e-05,
"loss": 1.2466,
"step": 995
},
{
"epoch": 1.0,
"learning_rate": 1.5727562017874602e-05,
"loss": 1.2547,
"step": 1000
},
{
"epoch": 1.0,
"eval_loss": 1.2699366807937622,
"eval_runtime": 152.6518,
"eval_samples_per_second": 92.708,
"eval_steps_per_second": 2.902,
"step": 1000
},
{
"epoch": 1.01,
"learning_rate": 1.5682721145969742e-05,
"loss": 1.2239,
"step": 1005
},
{
"epoch": 1.01,
"learning_rate": 1.5637710873327305e-05,
"loss": 1.228,
"step": 1010
},
{
"epoch": 1.02,
"learning_rate": 1.5592532541694173e-05,
"loss": 1.2122,
"step": 1015
},
{
"epoch": 1.02,
"learning_rate": 1.554718749782703e-05,
"loss": 1.2218,
"step": 1020
},
{
"epoch": 1.03,
"learning_rate": 1.5501677093452218e-05,
"loss": 1.228,
"step": 1025
},
{
"epoch": 1.03,
"learning_rate": 1.5456002685225447e-05,
"loss": 1.2189,
"step": 1030
},
{
"epoch": 1.04,
"learning_rate": 1.5410165634691343e-05,
"loss": 1.2289,
"step": 1035
},
{
"epoch": 1.04,
"learning_rate": 1.5364167308242866e-05,
"loss": 1.2214,
"step": 1040
},
{
"epoch": 1.05,
"learning_rate": 1.5318009077080577e-05,
"loss": 1.1938,
"step": 1045
},
{
"epoch": 1.05,
"learning_rate": 1.5271692317171765e-05,
"loss": 1.2181,
"step": 1050
},
{
"epoch": 1.06,
"learning_rate": 1.5225218409209424e-05,
"loss": 1.2311,
"step": 1055
},
{
"epoch": 1.06,
"learning_rate": 1.5178588738571108e-05,
"loss": 1.2348,
"step": 1060
},
{
"epoch": 1.07,
"learning_rate": 1.5131804695277612e-05,
"loss": 1.2268,
"step": 1065
},
{
"epoch": 1.07,
"learning_rate": 1.5084867673951558e-05,
"loss": 1.2187,
"step": 1070
},
{
"epoch": 1.08,
"learning_rate": 1.5037779073775805e-05,
"loss": 1.2285,
"step": 1075
},
{
"epoch": 1.08,
"learning_rate": 1.4990540298451748e-05,
"loss": 1.2053,
"step": 1080
},
{
"epoch": 1.09,
"learning_rate": 1.4943152756157476e-05,
"loss": 1.2075,
"step": 1085
},
{
"epoch": 1.09,
"learning_rate": 1.4895617859505782e-05,
"loss": 1.2191,
"step": 1090
},
{
"epoch": 1.1,
"learning_rate": 1.4847937025502067e-05,
"loss": 1.203,
"step": 1095
},
{
"epoch": 1.1,
"learning_rate": 1.4800111675502095e-05,
"loss": 1.2085,
"step": 1100
},
{
"epoch": 1.1,
"eval_loss": 1.2692837715148926,
"eval_runtime": 152.6613,
"eval_samples_per_second": 92.702,
"eval_steps_per_second": 2.902,
"step": 1100
},
{
"epoch": 1.11,
"learning_rate": 1.4752143235169623e-05,
"loss": 1.2413,
"step": 1105
},
{
"epoch": 1.11,
"learning_rate": 1.4704033134433895e-05,
"loss": 1.1931,
"step": 1110
},
{
"epoch": 1.12,
"learning_rate": 1.4655782807447025e-05,
"loss": 1.2152,
"step": 1115
},
{
"epoch": 1.12,
"learning_rate": 1.460739369254125e-05,
"loss": 1.2152,
"step": 1120
},
{
"epoch": 1.13,
"learning_rate": 1.4558867232186034e-05,
"loss": 1.2088,
"step": 1125
},
{
"epoch": 1.13,
"learning_rate": 1.4510204872945084e-05,
"loss": 1.1985,
"step": 1130
},
{
"epoch": 1.14,
"learning_rate": 1.446140806543323e-05,
"loss": 1.2047,
"step": 1135
},
{
"epoch": 1.14,
"learning_rate": 1.4412478264273173e-05,
"loss": 1.214,
"step": 1140
},
{
"epoch": 1.15,
"learning_rate": 1.4363416928052126e-05,
"loss": 1.222,
"step": 1145
},
{
"epoch": 1.15,
"learning_rate": 1.4314225519278333e-05,
"loss": 1.2188,
"step": 1150
},
{
"epoch": 1.16,
"learning_rate": 1.4264905504337483e-05,
"loss": 1.1926,
"step": 1155
},
{
"epoch": 1.16,
"learning_rate": 1.4215458353448981e-05,
"loss": 1.2045,
"step": 1160
},
{
"epoch": 1.17,
"learning_rate": 1.4165885540622128e-05,
"loss": 1.2075,
"step": 1165
},
{
"epoch": 1.17,
"learning_rate": 1.4116188543612182e-05,
"loss": 1.2222,
"step": 1170
},
{
"epoch": 1.18,
"learning_rate": 1.4066368843876305e-05,
"loss": 1.206,
"step": 1175
},
{
"epoch": 1.18,
"learning_rate": 1.4016427926529406e-05,
"loss": 1.2172,
"step": 1180
},
{
"epoch": 1.19,
"learning_rate": 1.3966367280299865e-05,
"loss": 1.2021,
"step": 1185
},
{
"epoch": 1.19,
"learning_rate": 1.3916188397485147e-05,
"loss": 1.2137,
"step": 1190
},
{
"epoch": 1.2,
"learning_rate": 1.3865892773907335e-05,
"loss": 1.2151,
"step": 1195
},
{
"epoch": 1.2,
"learning_rate": 1.3815481908868524e-05,
"loss": 1.2253,
"step": 1200
},
{
"epoch": 1.2,
"eval_loss": 1.266944408416748,
"eval_runtime": 152.7809,
"eval_samples_per_second": 92.629,
"eval_steps_per_second": 2.9,
"step": 1200
},
{
"epoch": 1.21,
"learning_rate": 1.3764957305106142e-05,
"loss": 1.2213,
"step": 1205
},
{
"epoch": 1.21,
"learning_rate": 1.3714320468748126e-05,
"loss": 1.2396,
"step": 1210
},
{
"epoch": 1.22,
"learning_rate": 1.3663572909268064e-05,
"loss": 1.195,
"step": 1215
},
{
"epoch": 1.22,
"learning_rate": 1.3612716139440164e-05,
"loss": 1.2225,
"step": 1220
},
{
"epoch": 1.23,
"learning_rate": 1.3561751675294182e-05,
"loss": 1.2257,
"step": 1225
},
{
"epoch": 1.23,
"learning_rate": 1.3510681036070206e-05,
"loss": 1.2274,
"step": 1230
},
{
"epoch": 1.24,
"learning_rate": 1.3459505744173392e-05,
"loss": 1.2203,
"step": 1235
},
{
"epoch": 1.24,
"learning_rate": 1.3408227325128569e-05,
"loss": 1.2354,
"step": 1240
},
{
"epoch": 1.25,
"learning_rate": 1.3356847307534759e-05,
"loss": 1.2147,
"step": 1245
},
{
"epoch": 1.25,
"learning_rate": 1.3305367223019622e-05,
"loss": 1.1968,
"step": 1250
},
{
"epoch": 1.26,
"learning_rate": 1.3253788606193791e-05,
"loss": 1.192,
"step": 1255
},
{
"epoch": 1.26,
"learning_rate": 1.3202112994605128e-05,
"loss": 1.1817,
"step": 1260
},
{
"epoch": 1.27,
"learning_rate": 1.3150341928692875e-05,
"loss": 1.207,
"step": 1265
},
{
"epoch": 1.27,
"learning_rate": 1.309847695174177e-05,
"loss": 1.1933,
"step": 1270
},
{
"epoch": 1.28,
"learning_rate": 1.3046519609836002e-05,
"loss": 1.2172,
"step": 1275
},
{
"epoch": 1.28,
"learning_rate": 1.2994471451813149e-05,
"loss": 1.2269,
"step": 1280
},
{
"epoch": 1.29,
"learning_rate": 1.2942334029217993e-05,
"loss": 1.2186,
"step": 1285
},
{
"epoch": 1.29,
"learning_rate": 1.2890108896256282e-05,
"loss": 1.2287,
"step": 1290
},
{
"epoch": 1.3,
"learning_rate": 1.2837797609748384e-05,
"loss": 1.2276,
"step": 1295
},
{
"epoch": 1.3,
"learning_rate": 1.2785401729082893e-05,
"loss": 1.215,
"step": 1300
},
{
"epoch": 1.3,
"eval_loss": 1.2649345397949219,
"eval_runtime": 152.6977,
"eval_samples_per_second": 92.68,
"eval_steps_per_second": 2.901,
"step": 1300
},
{
"epoch": 1.31,
"learning_rate": 1.273292281617013e-05,
"loss": 1.2203,
"step": 1305
},
{
"epoch": 1.31,
"learning_rate": 1.2680362435395595e-05,
"loss": 1.2208,
"step": 1310
},
{
"epoch": 1.32,
"learning_rate": 1.2627722153573326e-05,
"loss": 1.2206,
"step": 1315
},
{
"epoch": 1.32,
"learning_rate": 1.2575003539899198e-05,
"loss": 1.2156,
"step": 1320
},
{
"epoch": 1.33,
"learning_rate": 1.2522208165904137e-05,
"loss": 1.2144,
"step": 1325
},
{
"epoch": 1.33,
"learning_rate": 1.2469337605407284e-05,
"loss": 1.2217,
"step": 1330
},
{
"epoch": 1.34,
"learning_rate": 1.2416393434469068e-05,
"loss": 1.2111,
"step": 1335
},
{
"epoch": 1.34,
"learning_rate": 1.2363377231344239e-05,
"loss": 1.2218,
"step": 1340
},
{
"epoch": 1.35,
"learning_rate": 1.2310290576434795e-05,
"loss": 1.221,
"step": 1345
},
{
"epoch": 1.35,
"learning_rate": 1.2257135052242906e-05,
"loss": 1.2137,
"step": 1350
},
{
"epoch": 1.36,
"learning_rate": 1.220391224332371e-05,
"loss": 1.2183,
"step": 1355
},
{
"epoch": 1.36,
"learning_rate": 1.215062373623809e-05,
"loss": 1.2343,
"step": 1360
},
{
"epoch": 1.37,
"learning_rate": 1.2097271119505382e-05,
"loss": 1.2279,
"step": 1365
},
{
"epoch": 1.37,
"learning_rate": 1.2043855983556014e-05,
"loss": 1.2252,
"step": 1370
},
{
"epoch": 1.38,
"learning_rate": 1.1990379920684097e-05,
"loss": 1.2363,
"step": 1375
},
{
"epoch": 1.38,
"learning_rate": 1.1936844524999966e-05,
"loss": 1.2285,
"step": 1380
},
{
"epoch": 1.39,
"learning_rate": 1.1883251392382649e-05,
"loss": 1.2088,
"step": 1385
},
{
"epoch": 1.39,
"learning_rate": 1.1829602120432305e-05,
"loss": 1.2299,
"step": 1390
},
{
"epoch": 1.4,
"learning_rate": 1.177589830842259e-05,
"loss": 1.2032,
"step": 1395
},
{
"epoch": 1.4,
"learning_rate": 1.1722141557252998e-05,
"loss": 1.2103,
"step": 1400
},
{
"epoch": 1.4,
"eval_loss": 1.2630119323730469,
"eval_runtime": 152.7782,
"eval_samples_per_second": 92.631,
"eval_steps_per_second": 2.9,
"step": 1400
},
{
"epoch": 1.41,
"learning_rate": 1.1668333469401109e-05,
"loss": 1.2004,
"step": 1405
},
{
"epoch": 1.41,
"learning_rate": 1.1614475648874861e-05,
"loss": 1.2238,
"step": 1410
},
{
"epoch": 1.42,
"learning_rate": 1.1560569701164696e-05,
"loss": 1.2373,
"step": 1415
},
{
"epoch": 1.42,
"learning_rate": 1.1506617233195732e-05,
"loss": 1.2081,
"step": 1420
},
{
"epoch": 1.43,
"learning_rate": 1.145261985327983e-05,
"loss": 1.2074,
"step": 1425
},
{
"epoch": 1.43,
"learning_rate": 1.1398579171067679e-05,
"loss": 1.2049,
"step": 1430
},
{
"epoch": 1.44,
"learning_rate": 1.1344496797500795e-05,
"loss": 1.193,
"step": 1435
},
{
"epoch": 1.44,
"learning_rate": 1.1290374344763506e-05,
"loss": 1.2352,
"step": 1440
},
{
"epoch": 1.45,
"learning_rate": 1.1236213426234894e-05,
"loss": 1.1985,
"step": 1445
},
{
"epoch": 1.45,
"learning_rate": 1.1182015656440692e-05,
"loss": 1.1942,
"step": 1450
},
{
"epoch": 1.46,
"learning_rate": 1.112778265100517e-05,
"loss": 1.22,
"step": 1455
},
{
"epoch": 1.46,
"learning_rate": 1.1073516026602959e-05,
"loss": 1.2066,
"step": 1460
},
{
"epoch": 1.47,
"learning_rate": 1.1019217400910864e-05,
"loss": 1.1993,
"step": 1465
},
{
"epoch": 1.47,
"learning_rate": 1.096488839255965e-05,
"loss": 1.2043,
"step": 1470
},
{
"epoch": 1.48,
"learning_rate": 1.0910530621085769e-05,
"loss": 1.2347,
"step": 1475
},
{
"epoch": 1.48,
"learning_rate": 1.085614570688311e-05,
"loss": 1.2172,
"step": 1480
},
{
"epoch": 1.49,
"learning_rate": 1.080173527115467e-05,
"loss": 1.2095,
"step": 1485
},
{
"epoch": 1.49,
"learning_rate": 1.0747300935864245e-05,
"loss": 1.2196,
"step": 1490
},
{
"epoch": 1.5,
"learning_rate": 1.0692844323688065e-05,
"loss": 1.204,
"step": 1495
},
{
"epoch": 1.5,
"learning_rate": 1.0638367057966442e-05,
"loss": 1.2081,
"step": 1500
},
{
"epoch": 1.5,
"eval_loss": 1.261203646659851,
"eval_runtime": 152.8657,
"eval_samples_per_second": 92.578,
"eval_steps_per_second": 2.898,
"step": 1500
},
{
"epoch": 1.51,
"learning_rate": 1.0583870762655347e-05,
"loss": 1.1952,
"step": 1505
},
{
"epoch": 1.51,
"learning_rate": 1.0529357062278033e-05,
"loss": 1.2291,
"step": 1510
},
{
"epoch": 1.52,
"learning_rate": 1.0474827581876594e-05,
"loss": 1.2092,
"step": 1515
},
{
"epoch": 1.52,
"learning_rate": 1.042028394696352e-05,
"loss": 1.2042,
"step": 1520
},
{
"epoch": 1.53,
"learning_rate": 1.0365727783473238e-05,
"loss": 1.1966,
"step": 1525
},
{
"epoch": 1.53,
"learning_rate": 1.031116071771366e-05,
"loss": 1.2092,
"step": 1530
},
{
"epoch": 1.54,
"learning_rate": 1.0256584376317686e-05,
"loss": 1.2127,
"step": 1535
},
{
"epoch": 1.54,
"learning_rate": 1.0202000386194726e-05,
"loss": 1.2148,
"step": 1540
},
{
"epoch": 1.55,
"learning_rate": 1.0147410374482188e-05,
"loss": 1.1877,
"step": 1545
},
{
"epoch": 1.55,
"learning_rate": 1.009281596849699e-05,
"loss": 1.2095,
"step": 1550
},
{
"epoch": 1.56,
"learning_rate": 1.003821879568704e-05,
"loss": 1.1959,
"step": 1555
},
{
"epoch": 1.56,
"learning_rate": 9.983620483582726e-06,
"loss": 1.2015,
"step": 1560
},
{
"epoch": 1.57,
"learning_rate": 9.929022659748395e-06,
"loss": 1.2127,
"step": 1565
},
{
"epoch": 1.57,
"learning_rate": 9.874426951733844e-06,
"loss": 1.2026,
"step": 1570
},
{
"epoch": 1.58,
"learning_rate": 9.81983498702579e-06,
"loss": 1.2011,
"step": 1575
},
{
"epoch": 1.58,
"learning_rate": 9.76524839299937e-06,
"loss": 1.1925,
"step": 1580
},
{
"epoch": 1.59,
"learning_rate": 9.710668796869616e-06,
"loss": 1.2147,
"step": 1585
},
{
"epoch": 1.59,
"learning_rate": 9.65609782564296e-06,
"loss": 1.2214,
"step": 1590
},
{
"epoch": 1.6,
"learning_rate": 9.601537106068727e-06,
"loss": 1.1978,
"step": 1595
},
{
"epoch": 1.6,
"learning_rate": 9.546988264590637e-06,
"loss": 1.2033,
"step": 1600
},
{
"epoch": 1.6,
"eval_loss": 1.2596791982650757,
"eval_runtime": 152.6062,
"eval_samples_per_second": 92.735,
"eval_steps_per_second": 2.903,
"step": 1600
},
{
"epoch": 1.61,
"learning_rate": 9.49245292729833e-06,
"loss": 1.2225,
"step": 1605
},
{
"epoch": 1.61,
"learning_rate": 9.437932719878888e-06,
"loss": 1.2053,
"step": 1610
},
{
"epoch": 1.62,
"learning_rate": 9.383429267568379e-06,
"loss": 1.2083,
"step": 1615
},
{
"epoch": 1.62,
"learning_rate": 9.328944195103397e-06,
"loss": 1.2234,
"step": 1620
},
{
"epoch": 1.63,
"learning_rate": 9.27447912667264e-06,
"loss": 1.1946,
"step": 1625
},
{
"epoch": 1.63,
"learning_rate": 9.220035685868495e-06,
"loss": 1.2128,
"step": 1630
},
{
"epoch": 1.64,
"learning_rate": 9.165615495638627e-06,
"loss": 1.2111,
"step": 1635
},
{
"epoch": 1.64,
"learning_rate": 9.111220178237607e-06,
"loss": 1.2125,
"step": 1640
},
{
"epoch": 1.65,
"learning_rate": 9.056851355178556e-06,
"loss": 1.2026,
"step": 1645
},
{
"epoch": 1.65,
"learning_rate": 9.002510647184803e-06,
"loss": 1.2093,
"step": 1650
},
{
"epoch": 1.66,
"learning_rate": 8.94819967414157e-06,
"loss": 1.1912,
"step": 1655
},
{
"epoch": 1.66,
"learning_rate": 8.893920055047686e-06,
"loss": 1.201,
"step": 1660
},
{
"epoch": 1.67,
"learning_rate": 8.83967340796733e-06,
"loss": 1.203,
"step": 1665
},
{
"epoch": 1.67,
"learning_rate": 8.785461349981789e-06,
"loss": 1.2098,
"step": 1670
},
{
"epoch": 1.68,
"learning_rate": 8.731285497141258e-06,
"loss": 1.2068,
"step": 1675
},
{
"epoch": 1.68,
"learning_rate": 8.67714746441666e-06,
"loss": 1.2164,
"step": 1680
},
{
"epoch": 1.69,
"learning_rate": 8.62304886565151e-06,
"loss": 1.2234,
"step": 1685
},
{
"epoch": 1.69,
"learning_rate": 8.56899131351381e-06,
"loss": 1.2237,
"step": 1690
},
{
"epoch": 1.7,
"learning_rate": 8.514976419447963e-06,
"loss": 1.2053,
"step": 1695
},
{
"epoch": 1.7,
"learning_rate": 8.461005793626752e-06,
"loss": 1.2307,
"step": 1700
},
{
"epoch": 1.7,
"eval_loss": 1.2582082748413086,
"eval_runtime": 152.6347,
"eval_samples_per_second": 92.718,
"eval_steps_per_second": 2.902,
"step": 1700
},
{
"epoch": 1.71,
"learning_rate": 8.407081044903323e-06,
"loss": 1.1903,
"step": 1705
},
{
"epoch": 1.71,
"learning_rate": 8.353203780763239e-06,
"loss": 1.2058,
"step": 1710
},
{
"epoch": 1.72,
"learning_rate": 8.299375607276564e-06,
"loss": 1.2081,
"step": 1715
},
{
"epoch": 1.72,
"learning_rate": 8.245598129049973e-06,
"loss": 1.2275,
"step": 1720
},
{
"epoch": 1.73,
"learning_rate": 8.191872949178926e-06,
"loss": 1.2165,
"step": 1725
},
{
"epoch": 1.73,
"learning_rate": 8.138201669199878e-06,
"loss": 1.2105,
"step": 1730
},
{
"epoch": 1.74,
"learning_rate": 8.084585889042547e-06,
"loss": 1.1934,
"step": 1735
},
{
"epoch": 1.74,
"learning_rate": 8.031027206982211e-06,
"loss": 1.2045,
"step": 1740
},
{
"epoch": 1.75,
"learning_rate": 7.977527219592057e-06,
"loss": 1.2253,
"step": 1745
},
{
"epoch": 1.75,
"learning_rate": 7.924087521695602e-06,
"loss": 1.2139,
"step": 1750
},
{
"epoch": 1.76,
"learning_rate": 7.87070970631915e-06,
"loss": 1.2129,
"step": 1755
},
{
"epoch": 1.76,
"learning_rate": 7.817395364644294e-06,
"loss": 1.2117,
"step": 1760
},
{
"epoch": 1.77,
"learning_rate": 7.76414608596049e-06,
"loss": 1.2058,
"step": 1765
},
{
"epoch": 1.77,
"learning_rate": 7.710963457617681e-06,
"loss": 1.2286,
"step": 1770
},
{
"epoch": 1.78,
"learning_rate": 7.657849064978976e-06,
"loss": 1.2063,
"step": 1775
},
{
"epoch": 1.78,
"learning_rate": 7.604804491373396e-06,
"loss": 1.1878,
"step": 1780
},
{
"epoch": 1.79,
"learning_rate": 7.551831318048663e-06,
"loss": 1.1991,
"step": 1785
},
{
"epoch": 1.79,
"learning_rate": 7.498931124124077e-06,
"loss": 1.2007,
"step": 1790
},
{
"epoch": 1.8,
"learning_rate": 7.446105486543438e-06,
"loss": 1.2314,
"step": 1795
},
{
"epoch": 1.8,
"learning_rate": 7.393355980028039e-06,
"loss": 1.2038,
"step": 1800
},
{
"epoch": 1.8,
"eval_loss": 1.2567870616912842,
"eval_runtime": 152.7043,
"eval_samples_per_second": 92.676,
"eval_steps_per_second": 2.901,
"step": 1800
},
{
"epoch": 1.81,
"learning_rate": 7.340684177029709e-06,
"loss": 1.2296,
"step": 1805
},
{
"epoch": 1.81,
"learning_rate": 7.288091647683966e-06,
"loss": 1.2335,
"step": 1810
},
{
"epoch": 1.82,
"learning_rate": 7.235579959763189e-06,
"loss": 1.1986,
"step": 1815
},
{
"epoch": 1.82,
"learning_rate": 7.183150678629898e-06,
"loss": 1.1929,
"step": 1820
},
{
"epoch": 1.83,
"learning_rate": 7.1308053671900675e-06,
"loss": 1.2053,
"step": 1825
},
{
"epoch": 1.83,
"learning_rate": 7.078545585846578e-06,
"loss": 1.201,
"step": 1830
},
{
"epoch": 1.84,
"learning_rate": 7.026372892452653e-06,
"loss": 1.2166,
"step": 1835
},
{
"epoch": 1.84,
"learning_rate": 6.974288842265455e-06,
"loss": 1.2185,
"step": 1840
},
{
"epoch": 1.85,
"learning_rate": 6.922294987899705e-06,
"loss": 1.2157,
"step": 1845
},
{
"epoch": 1.85,
"learning_rate": 6.870392879281407e-06,
"loss": 1.2103,
"step": 1850
},
{
"epoch": 1.86,
"learning_rate": 6.818584063601646e-06,
"loss": 1.1871,
"step": 1855
},
{
"epoch": 1.86,
"learning_rate": 6.766870085270459e-06,
"loss": 1.2035,
"step": 1860
},
{
"epoch": 1.87,
"learning_rate": 6.715252485870804e-06,
"loss": 1.2021,
"step": 1865
},
{
"epoch": 1.87,
"learning_rate": 6.663732804112604e-06,
"loss": 1.1799,
"step": 1870
},
{
"epoch": 1.88,
"learning_rate": 6.6123125757868764e-06,
"loss": 1.1871,
"step": 1875
},
{
"epoch": 1.88,
"learning_rate": 6.5609933337199515e-06,
"loss": 1.1941,
"step": 1880
},
{
"epoch": 1.89,
"learning_rate": 6.50977660772778e-06,
"loss": 1.2048,
"step": 1885
},
{
"epoch": 1.89,
"learning_rate": 6.458663924570332e-06,
"loss": 1.2106,
"step": 1890
},
{
"epoch": 1.9,
"learning_rate": 6.4076568079060815e-06,
"loss": 1.2121,
"step": 1895
},
{
"epoch": 1.9,
"learning_rate": 6.356756778246587e-06,
"loss": 1.2014,
"step": 1900
},
{
"epoch": 1.9,
"eval_loss": 1.2556641101837158,
"eval_runtime": 152.825,
"eval_samples_per_second": 92.603,
"eval_steps_per_second": 2.899,
"step": 1900
},
{
"epoch": 1.91,
"learning_rate": 6.305965352911162e-06,
"loss": 1.2225,
"step": 1905
},
{
"epoch": 1.91,
"learning_rate": 6.255284045981653e-06,
"loss": 1.1912,
"step": 1910
},
{
"epoch": 1.92,
"learning_rate": 6.204714368257303e-06,
"loss": 1.2343,
"step": 1915
},
{
"epoch": 1.92,
"learning_rate": 6.1542578272096975e-06,
"loss": 1.2197,
"step": 1920
},
{
"epoch": 1.93,
"learning_rate": 6.103915926937853e-06,
"loss": 1.1902,
"step": 1925
},
{
"epoch": 1.93,
"learning_rate": 6.053690168123367e-06,
"loss": 1.2308,
"step": 1930
},
{
"epoch": 1.94,
"learning_rate": 6.003582047985684e-06,
"loss": 1.2306,
"step": 1935
},
{
"epoch": 1.94,
"learning_rate": 5.953593060237457e-06,
"loss": 1.2162,
"step": 1940
},
{
"epoch": 1.95,
"learning_rate": 5.903724695040036e-06,
"loss": 1.1785,
"step": 1945
},
{
"epoch": 1.95,
"learning_rate": 5.8539784389590345e-06,
"loss": 1.2002,
"step": 1950
},
{
"epoch": 1.96,
"learning_rate": 5.804355774920017e-06,
"loss": 1.2138,
"step": 1955
},
{
"epoch": 1.96,
"learning_rate": 5.754858182164298e-06,
"loss": 1.1786,
"step": 1960
},
{
"epoch": 1.97,
"learning_rate": 5.705487136204846e-06,
"loss": 1.223,
"step": 1965
},
{
"epoch": 1.97,
"learning_rate": 5.65624410878228e-06,
"loss": 1.2107,
"step": 1970
},
{
"epoch": 1.98,
"learning_rate": 5.60713056782103e-06,
"loss": 1.1922,
"step": 1975
},
{
"epoch": 1.98,
"learning_rate": 5.558147977385558e-06,
"loss": 1.2078,
"step": 1980
},
{
"epoch": 1.99,
"learning_rate": 5.509297797636712e-06,
"loss": 1.217,
"step": 1985
},
{
"epoch": 1.99,
"learning_rate": 5.46058148478821e-06,
"loss": 1.2227,
"step": 1990
},
{
"epoch": 2.0,
"learning_rate": 5.412000491063225e-06,
"loss": 1.2072,
"step": 1995
},
{
"epoch": 2.0,
"learning_rate": 5.363556264651093e-06,
"loss": 1.188,
"step": 2000
},
{
"epoch": 2.0,
"eval_loss": 1.2545795440673828,
"eval_runtime": 152.4763,
"eval_samples_per_second": 92.814,
"eval_steps_per_second": 2.905,
"step": 2000
},
{
"epoch": 2.01,
"learning_rate": 5.315250249664151e-06,
"loss": 1.1591,
"step": 2005
},
{
"epoch": 2.01,
"learning_rate": 5.267083886094668e-06,
"loss": 1.1676,
"step": 2010
},
{
"epoch": 2.02,
"learning_rate": 5.219058609771945e-06,
"loss": 1.1739,
"step": 2015
},
{
"epoch": 2.02,
"learning_rate": 5.171175852319503e-06,
"loss": 1.1684,
"step": 2020
},
{
"epoch": 2.03,
"learning_rate": 5.123437041112395e-06,
"loss": 1.1805,
"step": 2025
},
{
"epoch": 2.03,
"learning_rate": 5.075843599234678e-06,
"loss": 1.178,
"step": 2030
},
{
"epoch": 2.04,
"learning_rate": 5.0283969454369685e-06,
"loss": 1.1644,
"step": 2035
},
{
"epoch": 2.04,
"learning_rate": 4.98109849409417e-06,
"loss": 1.1745,
"step": 2040
},
{
"epoch": 2.05,
"learning_rate": 4.933949655163295e-06,
"loss": 1.1824,
"step": 2045
},
{
"epoch": 2.05,
"learning_rate": 4.886951834141446e-06,
"loss": 1.1768,
"step": 2050
},
{
"epoch": 2.06,
"learning_rate": 4.840106432023908e-06,
"loss": 1.1733,
"step": 2055
},
{
"epoch": 2.06,
"learning_rate": 4.7934148452623906e-06,
"loss": 1.17,
"step": 2060
},
{
"epoch": 2.07,
"learning_rate": 4.7468784657234055e-06,
"loss": 1.1679,
"step": 2065
},
{
"epoch": 2.07,
"learning_rate": 4.7004986806467635e-06,
"loss": 1.183,
"step": 2070
},
{
"epoch": 2.08,
"learning_rate": 4.65427687260423e-06,
"loss": 1.1705,
"step": 2075
},
{
"epoch": 2.08,
"learning_rate": 4.608214419458306e-06,
"loss": 1.1644,
"step": 2080
},
{
"epoch": 2.09,
"learning_rate": 4.562312694321156e-06,
"loss": 1.1739,
"step": 2085
},
{
"epoch": 2.09,
"learning_rate": 4.51657306551368e-06,
"loss": 1.1691,
"step": 2090
},
{
"epoch": 2.1,
"learning_rate": 4.470996896524713e-06,
"loss": 1.1825,
"step": 2095
},
{
"epoch": 2.1,
"learning_rate": 4.425585545970392e-06,
"loss": 1.1473,
"step": 2100
},
{
"epoch": 2.1,
"eval_loss": 1.2563046216964722,
"eval_runtime": 152.6039,
"eval_samples_per_second": 92.737,
"eval_steps_per_second": 2.903,
"step": 2100
},
{
"epoch": 2.11,
"learning_rate": 4.380340367553649e-06,
"loss": 1.2004,
"step": 2105
},
{
"epoch": 2.11,
"learning_rate": 4.335262710023861e-06,
"loss": 1.157,
"step": 2110
},
{
"epoch": 2.12,
"learning_rate": 4.290353917136639e-06,
"loss": 1.1748,
"step": 2115
},
{
"epoch": 2.12,
"learning_rate": 4.245615327613779e-06,
"loss": 1.1689,
"step": 2120
},
{
"epoch": 2.13,
"learning_rate": 4.201048275103339e-06,
"loss": 1.1691,
"step": 2125
},
{
"epoch": 2.13,
"learning_rate": 4.156654088139907e-06,
"loss": 1.1869,
"step": 2130
},
{
"epoch": 2.14,
"learning_rate": 4.112434090104978e-06,
"loss": 1.1709,
"step": 2135
},
{
"epoch": 2.14,
"learning_rate": 4.068389599187514e-06,
"loss": 1.175,
"step": 2140
},
{
"epoch": 2.15,
"learning_rate": 4.024521928344646e-06,
"loss": 1.1444,
"step": 2145
},
{
"epoch": 2.15,
"learning_rate": 3.980832385262532e-06,
"loss": 1.1824,
"step": 2150
},
{
"epoch": 2.16,
"learning_rate": 3.937322272317383e-06,
"loss": 1.1718,
"step": 2155
},
{
"epoch": 2.16,
"learning_rate": 3.893992886536637e-06,
"loss": 1.1699,
"step": 2160
},
{
"epoch": 2.17,
"learning_rate": 3.850845519560279e-06,
"loss": 1.1815,
"step": 2165
},
{
"epoch": 2.17,
"learning_rate": 3.807881457602367e-06,
"loss": 1.1679,
"step": 2170
},
{
"epoch": 2.18,
"learning_rate": 3.7651019814126656e-06,
"loss": 1.2018,
"step": 2175
},
{
"epoch": 2.18,
"learning_rate": 3.722508366238481e-06,
"loss": 1.1885,
"step": 2180
},
{
"epoch": 2.19,
"learning_rate": 3.6801018817866375e-06,
"loss": 1.1572,
"step": 2185
},
{
"epoch": 2.19,
"learning_rate": 3.6378837921856324e-06,
"loss": 1.1708,
"step": 2190
},
{
"epoch": 2.2,
"learning_rate": 3.5958553559479524e-06,
"loss": 1.1669,
"step": 2195
},
{
"epoch": 2.2,
"learning_rate": 3.5540178259325565e-06,
"loss": 1.1872,
"step": 2200
},
{
"epoch": 2.2,
"eval_loss": 1.2559261322021484,
"eval_runtime": 152.7815,
"eval_samples_per_second": 92.629,
"eval_steps_per_second": 2.9,
"step": 2200
},
{
"epoch": 2.21,
"learning_rate": 3.5123724493075318e-06,
"loss": 1.1767,
"step": 2205
},
{
"epoch": 2.21,
"learning_rate": 3.4709204675129025e-06,
"loss": 1.2005,
"step": 2210
},
{
"epoch": 2.22,
"learning_rate": 3.429663116223644e-06,
"loss": 1.1564,
"step": 2215
},
{
"epoch": 2.22,
"learning_rate": 3.388601625312833e-06,
"loss": 1.2063,
"step": 2220
},
{
"epoch": 2.23,
"learning_rate": 3.347737218814987e-06,
"loss": 1.1659,
"step": 2225
},
{
"epoch": 2.23,
"learning_rate": 3.307071114889583e-06,
"loss": 1.1777,
"step": 2230
},
{
"epoch": 2.24,
"learning_rate": 3.2666045257847332e-06,
"loss": 1.1724,
"step": 2235
},
{
"epoch": 2.24,
"learning_rate": 3.22633865780106e-06,
"loss": 1.1831,
"step": 2240
},
{
"epoch": 2.25,
"learning_rate": 3.1862747112557313e-06,
"loss": 1.171,
"step": 2245
},
{
"epoch": 2.25,
"learning_rate": 3.146413880446668e-06,
"loss": 1.183,
"step": 2250
},
{
"epoch": 2.26,
"learning_rate": 3.106757353616966e-06,
"loss": 1.1724,
"step": 2255
},
{
"epoch": 2.26,
"learning_rate": 3.067306312919456e-06,
"loss": 1.1855,
"step": 2260
},
{
"epoch": 2.27,
"learning_rate": 3.028061934381471e-06,
"loss": 1.169,
"step": 2265
},
{
"epoch": 2.27,
"learning_rate": 2.9890253878697885e-06,
"loss": 1.2027,
"step": 2270
},
{
"epoch": 2.28,
"learning_rate": 2.950197837055755e-06,
"loss": 1.1835,
"step": 2275
},
{
"epoch": 2.28,
"learning_rate": 2.911580439380599e-06,
"loss": 1.172,
"step": 2280
},
{
"epoch": 2.29,
"learning_rate": 2.8731743460209325e-06,
"loss": 1.1915,
"step": 2285
},
{
"epoch": 2.29,
"learning_rate": 2.834980701854417e-06,
"loss": 1.1736,
"step": 2290
},
{
"epoch": 2.3,
"learning_rate": 2.7970006454256605e-06,
"loss": 1.1861,
"step": 2295
},
{
"epoch": 2.3,
"learning_rate": 2.7592353089122582e-06,
"loss": 1.2086,
"step": 2300
},
{
"epoch": 2.3,
"eval_loss": 1.2553389072418213,
"eval_runtime": 152.643,
"eval_samples_per_second": 92.713,
"eval_steps_per_second": 2.902,
"step": 2300
},
{
"epoch": 2.31,
"learning_rate": 2.7216858180910532e-06,
"loss": 1.1664,
"step": 2305
},
{
"epoch": 2.31,
"learning_rate": 2.6843532923045702e-06,
"loss": 1.1633,
"step": 2310
},
{
"epoch": 2.32,
"learning_rate": 2.647238844427653e-06,
"loss": 1.1584,
"step": 2315
},
{
"epoch": 2.32,
"learning_rate": 2.610343580834287e-06,
"loss": 1.1762,
"step": 2320
},
{
"epoch": 2.33,
"learning_rate": 2.5736686013646226e-06,
"loss": 1.1907,
"step": 2325
},
{
"epoch": 2.33,
"learning_rate": 2.53721499929218e-06,
"loss": 1.1613,
"step": 2330
},
{
"epoch": 2.34,
"learning_rate": 2.5009838612912697e-06,
"loss": 1.153,
"step": 2335
},
{
"epoch": 2.34,
"learning_rate": 2.464976267404593e-06,
"loss": 1.1823,
"step": 2340
},
{
"epoch": 2.35,
"learning_rate": 2.4291932910110473e-06,
"loss": 1.164,
"step": 2345
},
{
"epoch": 2.35,
"learning_rate": 2.3936359987937353e-06,
"loss": 1.1889,
"step": 2350
},
{
"epoch": 2.36,
"learning_rate": 2.3583054507081526e-06,
"loss": 1.1655,
"step": 2355
},
{
"epoch": 2.36,
"learning_rate": 2.3232026999506062e-06,
"loss": 1.1608,
"step": 2360
},
{
"epoch": 2.37,
"learning_rate": 2.2883287929268027e-06,
"loss": 1.1638,
"step": 2365
},
{
"epoch": 2.37,
"learning_rate": 2.2536847692206722e-06,
"loss": 1.1982,
"step": 2370
},
{
"epoch": 2.38,
"learning_rate": 2.21927166156337e-06,
"loss": 1.1586,
"step": 2375
},
{
"epoch": 2.38,
"learning_rate": 2.185090495802491e-06,
"loss": 1.1639,
"step": 2380
},
{
"epoch": 2.39,
"learning_rate": 2.1511422908714897e-06,
"loss": 1.1899,
"step": 2385
},
{
"epoch": 2.39,
"learning_rate": 2.1174280587593076e-06,
"loss": 1.1679,
"step": 2390
},
{
"epoch": 2.4,
"learning_rate": 2.083948804480204e-06,
"loss": 1.1791,
"step": 2395
},
{
"epoch": 2.4,
"learning_rate": 2.0507055260438014e-06,
"loss": 1.1896,
"step": 2400
},
{
"epoch": 2.4,
"eval_loss": 1.25504469871521,
"eval_runtime": 152.7532,
"eval_samples_per_second": 92.646,
"eval_steps_per_second": 2.9,
"step": 2400
},
{
"epoch": 2.41,
"learning_rate": 2.017699214425323e-06,
"loss": 1.1863,
"step": 2405
},
{
"epoch": 2.41,
"learning_rate": 1.9849308535360688e-06,
"loss": 1.1658,
"step": 2410
},
{
"epoch": 2.42,
"learning_rate": 1.9524014201940732e-06,
"loss": 1.1743,
"step": 2415
},
{
"epoch": 2.42,
"learning_rate": 1.9201118840949905e-06,
"loss": 1.1695,
"step": 2420
},
{
"epoch": 2.43,
"learning_rate": 1.8880632077831906e-06,
"loss": 1.178,
"step": 2425
},
{
"epoch": 2.43,
"learning_rate": 1.8562563466230577e-06,
"loss": 1.1591,
"step": 2430
},
{
"epoch": 2.44,
"learning_rate": 1.824692248770521e-06,
"loss": 1.1913,
"step": 2435
},
{
"epoch": 2.44,
"learning_rate": 1.7933718551447877e-06,
"loss": 1.1801,
"step": 2440
},
{
"epoch": 2.45,
"learning_rate": 1.7622960994002836e-06,
"loss": 1.1923,
"step": 2445
},
{
"epoch": 2.45,
"learning_rate": 1.7314659078988383e-06,
"loss": 1.1584,
"step": 2450
},
{
"epoch": 2.46,
"learning_rate": 1.700882199682059e-06,
"loss": 1.174,
"step": 2455
},
{
"epoch": 2.46,
"learning_rate": 1.6705458864439405e-06,
"loss": 1.197,
"step": 2460
},
{
"epoch": 2.47,
"learning_rate": 1.6404578725036802e-06,
"loss": 1.1716,
"step": 2465
},
{
"epoch": 2.47,
"learning_rate": 1.6106190547787282e-06,
"loss": 1.1776,
"step": 2470
},
{
"epoch": 2.48,
"learning_rate": 1.5810303227580448e-06,
"loss": 1.1546,
"step": 2475
},
{
"epoch": 2.48,
"learning_rate": 1.5516925584755938e-06,
"loss": 1.1459,
"step": 2480
},
{
"epoch": 2.49,
"learning_rate": 1.5226066364840298e-06,
"loss": 1.1676,
"step": 2485
},
{
"epoch": 2.49,
"learning_rate": 1.4937734238286539e-06,
"loss": 1.1767,
"step": 2490
},
{
"epoch": 2.5,
"learning_rate": 1.4651937800215465e-06,
"loss": 1.1464,
"step": 2495
},
{
"epoch": 2.5,
"learning_rate": 1.436868557015959e-06,
"loss": 1.1733,
"step": 2500
},
{
"epoch": 2.5,
"eval_loss": 1.2547988891601562,
"eval_runtime": 152.7044,
"eval_samples_per_second": 92.676,
"eval_steps_per_second": 2.901,
"step": 2500
},
{
"epoch": 2.51,
"learning_rate": 1.4087985991809072e-06,
"loss": 1.1714,
"step": 2505
},
{
"epoch": 2.51,
"learning_rate": 1.380984743276006e-06,
"loss": 1.1802,
"step": 2510
},
{
"epoch": 2.52,
"learning_rate": 1.353427818426527e-06,
"loss": 1.174,
"step": 2515
},
{
"epoch": 2.52,
"learning_rate": 1.32612864609868e-06,
"loss": 1.1993,
"step": 2520
},
{
"epoch": 2.53,
"learning_rate": 1.2990880400751204e-06,
"loss": 1.1689,
"step": 2525
},
{
"epoch": 2.53,
"learning_rate": 1.272306806430702e-06,
"loss": 1.1681,
"step": 2530
},
{
"epoch": 2.54,
"learning_rate": 1.245785743508441e-06,
"loss": 1.1841,
"step": 2535
},
{
"epoch": 2.54,
"learning_rate": 1.2195256418957146e-06,
"loss": 1.1767,
"step": 2540
},
{
"epoch": 2.55,
"learning_rate": 1.1935272844007029e-06,
"loss": 1.1753,
"step": 2545
},
{
"epoch": 2.55,
"learning_rate": 1.167791446029043e-06,
"loss": 1.1547,
"step": 2550
},
{
"epoch": 2.56,
"learning_rate": 1.1423188939607356e-06,
"loss": 1.2079,
"step": 2555
},
{
"epoch": 2.56,
"learning_rate": 1.1171103875272693e-06,
"loss": 1.1776,
"step": 2560
},
{
"epoch": 2.57,
"learning_rate": 1.0921666781889861e-06,
"loss": 1.192,
"step": 2565
},
{
"epoch": 2.57,
"learning_rate": 1.067488509512683e-06,
"loss": 1.1615,
"step": 2570
},
{
"epoch": 2.58,
"learning_rate": 1.0430766171494444e-06,
"loss": 1.1664,
"step": 2575
},
{
"epoch": 2.58,
"learning_rate": 1.0189317288127154e-06,
"loss": 1.1749,
"step": 2580
},
{
"epoch": 2.59,
"learning_rate": 9.950545642566045e-07,
"loss": 1.1912,
"step": 2585
},
{
"epoch": 2.59,
"learning_rate": 9.714458352544276e-07,
"loss": 1.1704,
"step": 2590
},
{
"epoch": 2.6,
"learning_rate": 9.481062455774981e-07,
"loss": 1.1674,
"step": 2595
},
{
"epoch": 2.6,
"learning_rate": 9.25036490974136e-07,
"loss": 1.1665,
"step": 2600
},
{
"epoch": 2.6,
"eval_loss": 1.2544310092926025,
"eval_runtime": 152.8108,
"eval_samples_per_second": 92.611,
"eval_steps_per_second": 2.899,
"step": 2600
},
{
"epoch": 2.61,
"learning_rate": 9.022372591489381e-07,
"loss": 1.1531,
"step": 2605
},
{
"epoch": 2.61,
"learning_rate": 8.797092297422715e-07,
"loss": 1.191,
"step": 2610
},
{
"epoch": 2.62,
"learning_rate": 8.574530743100151e-07,
"loss": 1.1529,
"step": 2615
},
{
"epoch": 2.62,
"learning_rate": 8.354694563035426e-07,
"loss": 1.1724,
"step": 2620
},
{
"epoch": 2.63,
"learning_rate": 8.137590310499433e-07,
"loss": 1.1698,
"step": 2625
},
{
"epoch": 2.63,
"learning_rate": 7.923224457324852e-07,
"loss": 1.1841,
"step": 2630
},
{
"epoch": 2.64,
"learning_rate": 7.711603393713297e-07,
"loss": 1.1679,
"step": 2635
},
{
"epoch": 2.64,
"learning_rate": 7.502733428044684e-07,
"loss": 1.185,
"step": 2640
},
{
"epoch": 2.65,
"learning_rate": 7.296620786689356e-07,
"loss": 1.1784,
"step": 2645
},
{
"epoch": 2.65,
"learning_rate": 7.093271613822373e-07,
"loss": 1.1703,
"step": 2650
},
{
"epoch": 2.66,
"learning_rate": 6.89269197124034e-07,
"loss": 1.1825,
"step": 2655
},
{
"epoch": 2.66,
"learning_rate": 6.694887838180764e-07,
"loss": 1.1789,
"step": 2660
},
{
"epoch": 2.67,
"learning_rate": 6.499865111143788e-07,
"loss": 1.1693,
"step": 2665
},
{
"epoch": 2.67,
"learning_rate": 6.307629603716425e-07,
"loss": 1.1682,
"step": 2670
},
{
"epoch": 2.68,
"learning_rate": 6.11818704639926e-07,
"loss": 1.1778,
"step": 2675
},
{
"epoch": 2.68,
"learning_rate": 5.931543086435543e-07,
"loss": 1.1951,
"step": 2680
},
{
"epoch": 2.69,
"learning_rate": 5.747703287643003e-07,
"loss": 1.1818,
"step": 2685
},
{
"epoch": 2.69,
"learning_rate": 5.566673130247879e-07,
"loss": 1.1722,
"step": 2690
},
{
"epoch": 2.7,
"learning_rate": 5.388458010721575e-07,
"loss": 1.1669,
"step": 2695
},
{
"epoch": 2.7,
"learning_rate": 5.213063241619798e-07,
"loss": 1.1499,
"step": 2700
},
{
"epoch": 2.7,
"eval_loss": 1.254270076751709,
"eval_runtime": 152.9074,
"eval_samples_per_second": 92.553,
"eval_steps_per_second": 2.897,
"step": 2700
},
{
"epoch": 2.71,
"learning_rate": 5.040494051424205e-07,
"loss": 1.172,
"step": 2705
},
{
"epoch": 2.71,
"learning_rate": 4.870755584386544e-07,
"loss": 1.1771,
"step": 2710
},
{
"epoch": 2.72,
"learning_rate": 4.703852900375272e-07,
"loss": 1.1574,
"step": 2715
},
{
"epoch": 2.72,
"learning_rate": 4.539790974724723e-07,
"loss": 1.178,
"step": 2720
},
{
"epoch": 2.73,
"learning_rate": 4.378574698086868e-07,
"loss": 1.1696,
"step": 2725
},
{
"epoch": 2.73,
"learning_rate": 4.2202088762854255e-07,
"loss": 1.1794,
"step": 2730
},
{
"epoch": 2.74,
"learning_rate": 4.064698230172681e-07,
"loss": 1.17,
"step": 2735
},
{
"epoch": 2.74,
"learning_rate": 3.912047395488705e-07,
"loss": 1.1873,
"step": 2740
},
{
"epoch": 2.75,
"learning_rate": 3.762260922723182e-07,
"loss": 1.1786,
"step": 2745
},
{
"epoch": 2.75,
"learning_rate": 3.615343276979777e-07,
"loss": 1.1958,
"step": 2750
},
{
"epoch": 2.76,
"learning_rate": 3.471298837843007e-07,
"loss": 1.1831,
"step": 2755
},
{
"epoch": 2.76,
"learning_rate": 3.3301318992476904e-07,
"loss": 1.1855,
"step": 2760
},
{
"epoch": 2.77,
"learning_rate": 3.191846669350973e-07,
"loss": 1.1659,
"step": 2765
},
{
"epoch": 2.77,
"learning_rate": 3.056447270406826e-07,
"loss": 1.1666,
"step": 2770
},
{
"epoch": 2.78,
"learning_rate": 2.9239377386432345e-07,
"loss": 1.172,
"step": 2775
},
{
"epoch": 2.78,
"learning_rate": 2.7943220241418376e-07,
"loss": 1.1759,
"step": 2780
},
{
"epoch": 2.79,
"learning_rate": 2.667603990720169e-07,
"loss": 1.1766,
"step": 2785
},
{
"epoch": 2.79,
"learning_rate": 2.5437874158164897e-07,
"loss": 1.1744,
"step": 2790
},
{
"epoch": 2.8,
"learning_rate": 2.422875990377205e-07,
"loss": 1.2014,
"step": 2795
},
{
"epoch": 2.8,
"learning_rate": 2.3048733187468032e-07,
"loss": 1.1779,
"step": 2800
},
{
"epoch": 2.8,
"eval_loss": 1.2542082071304321,
"eval_runtime": 152.7562,
"eval_samples_per_second": 92.644,
"eval_steps_per_second": 2.9,
"step": 2800
},
{
"epoch": 2.81,
"learning_rate": 2.1897829185604235e-07,
"loss": 1.1756,
"step": 2805
},
{
"epoch": 2.81,
"learning_rate": 2.0776082206389936e-07,
"loss": 1.1919,
"step": 2810
},
{
"epoch": 2.82,
"learning_rate": 1.9683525688869776e-07,
"loss": 1.1807,
"step": 2815
},
{
"epoch": 2.82,
"learning_rate": 1.862019220192679e-07,
"loss": 1.1878,
"step": 2820
},
{
"epoch": 2.83,
"learning_rate": 1.7586113443311404e-07,
"loss": 1.1695,
"step": 2825
},
{
"epoch": 2.83,
"learning_rate": 1.658132023869663e-07,
"loss": 1.1949,
"step": 2830
},
{
"epoch": 2.84,
"learning_rate": 1.5605842540759476e-07,
"loss": 1.1779,
"step": 2835
},
{
"epoch": 2.84,
"learning_rate": 1.4659709428287538e-07,
"loss": 1.177,
"step": 2840
},
{
"epoch": 2.85,
"learning_rate": 1.3742949105312375e-07,
"loss": 1.1727,
"step": 2845
},
{
"epoch": 2.85,
"learning_rate": 1.2855588900269057e-07,
"loss": 1.1921,
"step": 2850
},
{
"epoch": 2.86,
"learning_rate": 1.199765526518082e-07,
"loss": 1.1768,
"step": 2855
},
{
"epoch": 2.86,
"learning_rate": 1.1169173774871478e-07,
"loss": 1.1775,
"step": 2860
},
{
"epoch": 2.87,
"learning_rate": 1.0370169126202145e-07,
"loss": 1.1533,
"step": 2865
},
{
"epoch": 2.87,
"learning_rate": 9.600665137335374e-08,
"loss": 1.202,
"step": 2870
},
{
"epoch": 2.88,
"learning_rate": 8.860684747025727e-08,
"loss": 1.1789,
"step": 2875
},
{
"epoch": 2.88,
"learning_rate": 8.150250013934547e-08,
"loss": 1.1725,
"step": 2880
},
{
"epoch": 2.89,
"learning_rate": 7.469382115974034e-08,
"loss": 1.1604,
"step": 2885
},
{
"epoch": 2.89,
"learning_rate": 6.818101349674756e-08,
"loss": 1.1795,
"step": 2890
},
{
"epoch": 2.9,
"learning_rate": 6.196427129581129e-08,
"loss": 1.1658,
"step": 2895
},
{
"epoch": 2.9,
"learning_rate": 5.60437798767266e-08,
"loss": 1.1746,
"step": 2900
},
{
"epoch": 2.9,
"eval_loss": 1.254175066947937,
"eval_runtime": 152.7765,
"eval_samples_per_second": 92.632,
"eval_steps_per_second": 2.9,
"step": 2900
},
{
"epoch": 2.91,
"learning_rate": 5.041971572811277e-08,
"loss": 1.1694,
"step": 2905
},
{
"epoch": 2.91,
"learning_rate": 4.509224650215527e-08,
"loss": 1.192,
"step": 2910
},
{
"epoch": 2.92,
"learning_rate": 4.0061531009605305e-08,
"loss": 1.1549,
"step": 2915
},
{
"epoch": 2.92,
"learning_rate": 3.5327719215046965e-08,
"loss": 1.1779,
"step": 2920
},
{
"epoch": 2.93,
"learning_rate": 3.089095223242744e-08,
"loss": 1.1734,
"step": 2925
},
{
"epoch": 2.93,
"learning_rate": 2.6751362320849293e-08,
"loss": 1.161,
"step": 2930
},
{
"epoch": 2.94,
"learning_rate": 2.2909072880629156e-08,
"loss": 1.1553,
"step": 2935
},
{
"epoch": 2.94,
"learning_rate": 1.9364198449616233e-08,
"loss": 1.1885,
"step": 2940
},
{
"epoch": 2.95,
"learning_rate": 1.6116844699781697e-08,
"loss": 1.1814,
"step": 2945
},
{
"epoch": 2.95,
"learning_rate": 1.3167108434066766e-08,
"loss": 1.1724,
"step": 2950
},
{
"epoch": 2.96,
"learning_rate": 1.0515077583498346e-08,
"loss": 1.1942,
"step": 2955
},
{
"epoch": 2.96,
"learning_rate": 8.160831204563347e-09,
"loss": 1.207,
"step": 2960
},
{
"epoch": 2.97,
"learning_rate": 6.1044394768594585e-09,
"loss": 1.1665,
"step": 2965
},
{
"epoch": 2.97,
"learning_rate": 4.345963700995714e-09,
"loss": 1.1608,
"step": 2970
},
{
"epoch": 2.98,
"learning_rate": 2.8854562967706167e-09,
"loss": 1.182,
"step": 2975
},
{
"epoch": 2.98,
"learning_rate": 1.7229608016067213e-09,
"loss": 1.1413,
"step": 2980
},
{
"epoch": 2.99,
"learning_rate": 8.585118692538974e-10,
"loss": 1.1668,
"step": 2985
},
{
"epoch": 2.99,
"learning_rate": 2.9213526875349063e-10,
"loss": 1.1881,
"step": 2990
},
{
"epoch": 3.0,
"learning_rate": 2.384788367670865e-11,
"loss": 1.1883,
"step": 2995
},
{
"epoch": 3.0,
"step": 2997,
"total_flos": 6.191263180086313e+18,
"train_loss": 1.2270306770030681,
"train_runtime": 19334.7166,
"train_samples_per_second": 19.844,
"train_steps_per_second": 0.155
}
],
"logging_steps": 5,
"max_steps": 2997,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 100,
"total_flos": 6.191263180086313e+18,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}