generated from xuyuqing/ailab
2106 lines
50 KiB
JSON
2106 lines
50 KiB
JSON
{
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 200.0,
|
|
"global_step": 800,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"epoch": 1.0,
|
|
"eval_loss": 2.419327735900879,
|
|
"eval_runtime": 0.2863,
|
|
"eval_samples_per_second": 433.051,
|
|
"eval_steps_per_second": 13.969,
|
|
"step": 4
|
|
},
|
|
{
|
|
"epoch": 2.0,
|
|
"eval_loss": 2.3889026641845703,
|
|
"eval_runtime": 0.2828,
|
|
"eval_samples_per_second": 438.427,
|
|
"eval_steps_per_second": 14.143,
|
|
"step": 8
|
|
},
|
|
{
|
|
"epoch": 2.5,
|
|
"learning_rate": 2.0833333333333336e-05,
|
|
"loss": 2.3682,
|
|
"step": 10
|
|
},
|
|
{
|
|
"epoch": 3.0,
|
|
"eval_loss": 2.3414814472198486,
|
|
"eval_runtime": 0.2841,
|
|
"eval_samples_per_second": 436.463,
|
|
"eval_steps_per_second": 14.079,
|
|
"step": 12
|
|
},
|
|
{
|
|
"epoch": 4.0,
|
|
"eval_loss": 2.275804042816162,
|
|
"eval_runtime": 0.2868,
|
|
"eval_samples_per_second": 432.329,
|
|
"eval_steps_per_second": 13.946,
|
|
"step": 16
|
|
},
|
|
{
|
|
"epoch": 5.0,
|
|
"learning_rate": 4.166666666666667e-05,
|
|
"loss": 2.288,
|
|
"step": 20
|
|
},
|
|
{
|
|
"epoch": 5.0,
|
|
"eval_loss": 2.182788372039795,
|
|
"eval_runtime": 0.2816,
|
|
"eval_samples_per_second": 440.276,
|
|
"eval_steps_per_second": 14.202,
|
|
"step": 20
|
|
},
|
|
{
|
|
"epoch": 6.0,
|
|
"eval_loss": 2.082679271697998,
|
|
"eval_runtime": 0.2786,
|
|
"eval_samples_per_second": 445.107,
|
|
"eval_steps_per_second": 14.358,
|
|
"step": 24
|
|
},
|
|
{
|
|
"epoch": 7.0,
|
|
"eval_loss": 2.0054304599761963,
|
|
"eval_runtime": 0.2806,
|
|
"eval_samples_per_second": 441.844,
|
|
"eval_steps_per_second": 14.253,
|
|
"step": 28
|
|
},
|
|
{
|
|
"epoch": 7.5,
|
|
"learning_rate": 4.99926249076577e-05,
|
|
"loss": 2.0593,
|
|
"step": 30
|
|
},
|
|
{
|
|
"epoch": 8.0,
|
|
"eval_loss": 1.9229722023010254,
|
|
"eval_runtime": 0.2819,
|
|
"eval_samples_per_second": 439.868,
|
|
"eval_steps_per_second": 14.189,
|
|
"step": 32
|
|
},
|
|
{
|
|
"epoch": 9.0,
|
|
"eval_loss": 1.852725625038147,
|
|
"eval_runtime": 0.2826,
|
|
"eval_samples_per_second": 438.746,
|
|
"eval_steps_per_second": 14.153,
|
|
"step": 36
|
|
},
|
|
{
|
|
"epoch": 10.0,
|
|
"learning_rate": 4.9947570655942796e-05,
|
|
"loss": 1.8578,
|
|
"step": 40
|
|
},
|
|
{
|
|
"epoch": 10.0,
|
|
"eval_loss": 1.7921116352081299,
|
|
"eval_runtime": 0.2803,
|
|
"eval_samples_per_second": 442.428,
|
|
"eval_steps_per_second": 14.272,
|
|
"step": 40
|
|
},
|
|
{
|
|
"epoch": 11.0,
|
|
"eval_loss": 1.7393769025802612,
|
|
"eval_runtime": 0.2803,
|
|
"eval_samples_per_second": 442.33,
|
|
"eval_steps_per_second": 14.269,
|
|
"step": 44
|
|
},
|
|
{
|
|
"epoch": 12.0,
|
|
"eval_loss": 1.689049243927002,
|
|
"eval_runtime": 0.278,
|
|
"eval_samples_per_second": 445.972,
|
|
"eval_steps_per_second": 14.386,
|
|
"step": 48
|
|
},
|
|
{
|
|
"epoch": 12.5,
|
|
"learning_rate": 4.986163316987876e-05,
|
|
"loss": 1.688,
|
|
"step": 50
|
|
},
|
|
{
|
|
"epoch": 13.0,
|
|
"eval_loss": 1.6388040781021118,
|
|
"eval_runtime": 0.2817,
|
|
"eval_samples_per_second": 440.188,
|
|
"eval_steps_per_second": 14.2,
|
|
"step": 52
|
|
},
|
|
{
|
|
"epoch": 14.0,
|
|
"eval_loss": 1.5914537906646729,
|
|
"eval_runtime": 0.2822,
|
|
"eval_samples_per_second": 439.33,
|
|
"eval_steps_per_second": 14.172,
|
|
"step": 56
|
|
},
|
|
{
|
|
"epoch": 15.0,
|
|
"learning_rate": 4.9734953280908904e-05,
|
|
"loss": 1.5634,
|
|
"step": 60
|
|
},
|
|
{
|
|
"epoch": 15.0,
|
|
"eval_loss": 1.5468162298202515,
|
|
"eval_runtime": 0.2824,
|
|
"eval_samples_per_second": 439.059,
|
|
"eval_steps_per_second": 14.163,
|
|
"step": 60
|
|
},
|
|
{
|
|
"epoch": 16.0,
|
|
"eval_loss": 1.5048052072525024,
|
|
"eval_runtime": 0.2816,
|
|
"eval_samples_per_second": 440.291,
|
|
"eval_steps_per_second": 14.203,
|
|
"step": 64
|
|
},
|
|
{
|
|
"epoch": 17.0,
|
|
"eval_loss": 1.4672209024429321,
|
|
"eval_runtime": 0.314,
|
|
"eval_samples_per_second": 394.897,
|
|
"eval_steps_per_second": 12.739,
|
|
"step": 68
|
|
},
|
|
{
|
|
"epoch": 17.5,
|
|
"learning_rate": 4.956773858774731e-05,
|
|
"loss": 1.4744,
|
|
"step": 70
|
|
},
|
|
{
|
|
"epoch": 18.0,
|
|
"eval_loss": 1.4316694736480713,
|
|
"eval_runtime": 0.2805,
|
|
"eval_samples_per_second": 442.088,
|
|
"eval_steps_per_second": 14.261,
|
|
"step": 72
|
|
},
|
|
{
|
|
"epoch": 19.0,
|
|
"eval_loss": 1.3966118097305298,
|
|
"eval_runtime": 0.2873,
|
|
"eval_samples_per_second": 431.642,
|
|
"eval_steps_per_second": 13.924,
|
|
"step": 76
|
|
},
|
|
{
|
|
"epoch": 20.0,
|
|
"learning_rate": 4.936026311617316e-05,
|
|
"loss": 1.4126,
|
|
"step": 80
|
|
},
|
|
{
|
|
"epoch": 20.0,
|
|
"eval_loss": 1.361703634262085,
|
|
"eval_runtime": 0.2798,
|
|
"eval_samples_per_second": 443.096,
|
|
"eval_steps_per_second": 14.293,
|
|
"step": 80
|
|
},
|
|
{
|
|
"epoch": 21.0,
|
|
"eval_loss": 1.3309640884399414,
|
|
"eval_runtime": 0.2799,
|
|
"eval_samples_per_second": 443.011,
|
|
"eval_steps_per_second": 14.291,
|
|
"step": 84
|
|
},
|
|
{
|
|
"epoch": 22.0,
|
|
"eval_loss": 1.3007252216339111,
|
|
"eval_runtime": 0.2799,
|
|
"eval_samples_per_second": 443.029,
|
|
"eval_steps_per_second": 14.291,
|
|
"step": 88
|
|
},
|
|
{
|
|
"epoch": 22.5,
|
|
"learning_rate": 4.9112866869966475e-05,
|
|
"loss": 1.3297,
|
|
"step": 90
|
|
},
|
|
{
|
|
"epoch": 23.0,
|
|
"eval_loss": 1.2726101875305176,
|
|
"eval_runtime": 0.2807,
|
|
"eval_samples_per_second": 441.683,
|
|
"eval_steps_per_second": 14.248,
|
|
"step": 92
|
|
},
|
|
{
|
|
"epoch": 24.0,
|
|
"eval_loss": 1.2463679313659668,
|
|
"eval_runtime": 0.2797,
|
|
"eval_samples_per_second": 443.37,
|
|
"eval_steps_per_second": 14.302,
|
|
"step": 96
|
|
},
|
|
{
|
|
"epoch": 25.0,
|
|
"learning_rate": 4.882595527372152e-05,
|
|
"loss": 1.2377,
|
|
"step": 100
|
|
},
|
|
{
|
|
"epoch": 25.0,
|
|
"eval_loss": 1.2251466512680054,
|
|
"eval_runtime": 0.281,
|
|
"eval_samples_per_second": 441.285,
|
|
"eval_steps_per_second": 14.235,
|
|
"step": 100
|
|
},
|
|
{
|
|
"epoch": 26.0,
|
|
"eval_loss": 1.1965230703353882,
|
|
"eval_runtime": 0.2811,
|
|
"eval_samples_per_second": 441.079,
|
|
"eval_steps_per_second": 14.228,
|
|
"step": 104
|
|
},
|
|
{
|
|
"epoch": 27.0,
|
|
"eval_loss": 1.170884370803833,
|
|
"eval_runtime": 0.2806,
|
|
"eval_samples_per_second": 441.989,
|
|
"eval_steps_per_second": 14.258,
|
|
"step": 108
|
|
},
|
|
{
|
|
"epoch": 27.5,
|
|
"learning_rate": 4.8499998508450664e-05,
|
|
"loss": 1.1946,
|
|
"step": 110
|
|
},
|
|
{
|
|
"epoch": 28.0,
|
|
"eval_loss": 1.1476936340332031,
|
|
"eval_runtime": 0.2857,
|
|
"eval_samples_per_second": 434.039,
|
|
"eval_steps_per_second": 14.001,
|
|
"step": 112
|
|
},
|
|
{
|
|
"epoch": 29.0,
|
|
"eval_loss": 1.1263080835342407,
|
|
"eval_runtime": 0.2803,
|
|
"eval_samples_per_second": 442.444,
|
|
"eval_steps_per_second": 14.272,
|
|
"step": 116
|
|
},
|
|
{
|
|
"epoch": 30.0,
|
|
"learning_rate": 4.813553074106761e-05,
|
|
"loss": 1.1089,
|
|
"step": 120
|
|
},
|
|
{
|
|
"epoch": 30.0,
|
|
"eval_loss": 1.1085485219955444,
|
|
"eval_runtime": 0.2804,
|
|
"eval_samples_per_second": 442.238,
|
|
"eval_steps_per_second": 14.266,
|
|
"step": 120
|
|
},
|
|
{
|
|
"epoch": 31.0,
|
|
"eval_loss": 1.0873397588729858,
|
|
"eval_runtime": 0.2899,
|
|
"eval_samples_per_second": 427.672,
|
|
"eval_steps_per_second": 13.796,
|
|
"step": 124
|
|
},
|
|
{
|
|
"epoch": 32.0,
|
|
"eval_loss": 1.0683445930480957,
|
|
"eval_runtime": 0.2811,
|
|
"eval_samples_per_second": 441.113,
|
|
"eval_steps_per_second": 14.229,
|
|
"step": 128
|
|
},
|
|
{
|
|
"epoch": 32.5,
|
|
"learning_rate": 4.773314924901281e-05,
|
|
"loss": 1.073,
|
|
"step": 130
|
|
},
|
|
{
|
|
"epoch": 33.0,
|
|
"eval_loss": 1.0528886318206787,
|
|
"eval_runtime": 0.2812,
|
|
"eval_samples_per_second": 440.97,
|
|
"eval_steps_per_second": 14.225,
|
|
"step": 132
|
|
},
|
|
{
|
|
"epoch": 34.0,
|
|
"eval_loss": 1.0351759195327759,
|
|
"eval_runtime": 0.2804,
|
|
"eval_samples_per_second": 442.186,
|
|
"eval_steps_per_second": 14.264,
|
|
"step": 136
|
|
},
|
|
{
|
|
"epoch": 35.0,
|
|
"learning_rate": 4.7293513441455364e-05,
|
|
"loss": 1.0076,
|
|
"step": 140
|
|
},
|
|
{
|
|
"epoch": 35.0,
|
|
"eval_loss": 1.01103937625885,
|
|
"eval_runtime": 0.2805,
|
|
"eval_samples_per_second": 442.046,
|
|
"eval_steps_per_second": 14.26,
|
|
"step": 140
|
|
},
|
|
{
|
|
"epoch": 36.0,
|
|
"eval_loss": 0.9903552532196045,
|
|
"eval_runtime": 0.2818,
|
|
"eval_samples_per_second": 440.023,
|
|
"eval_steps_per_second": 14.194,
|
|
"step": 144
|
|
},
|
|
{
|
|
"epoch": 37.0,
|
|
"eval_loss": 0.9713702201843262,
|
|
"eval_runtime": 0.2792,
|
|
"eval_samples_per_second": 444.166,
|
|
"eval_steps_per_second": 14.328,
|
|
"step": 148
|
|
},
|
|
{
|
|
"epoch": 37.5,
|
|
"learning_rate": 4.6817343778675614e-05,
|
|
"loss": 0.9457,
|
|
"step": 150
|
|
},
|
|
{
|
|
"epoch": 38.0,
|
|
"eval_loss": 0.9592961668968201,
|
|
"eval_runtime": 0.2835,
|
|
"eval_samples_per_second": 437.423,
|
|
"eval_steps_per_second": 14.11,
|
|
"step": 152
|
|
},
|
|
{
|
|
"epoch": 39.0,
|
|
"eval_loss": 0.9450886249542236,
|
|
"eval_runtime": 0.2807,
|
|
"eval_samples_per_second": 441.817,
|
|
"eval_steps_per_second": 14.252,
|
|
"step": 156
|
|
},
|
|
{
|
|
"epoch": 40.0,
|
|
"learning_rate": 4.630542059139924e-05,
|
|
"loss": 0.8931,
|
|
"step": 160
|
|
},
|
|
{
|
|
"epoch": 40.0,
|
|
"eval_loss": 0.922899603843689,
|
|
"eval_runtime": 0.2816,
|
|
"eval_samples_per_second": 440.302,
|
|
"eval_steps_per_second": 14.203,
|
|
"step": 160
|
|
},
|
|
{
|
|
"epoch": 41.0,
|
|
"eval_loss": 0.9055746793746948,
|
|
"eval_runtime": 0.2958,
|
|
"eval_samples_per_second": 419.191,
|
|
"eval_steps_per_second": 13.522,
|
|
"step": 164
|
|
},
|
|
{
|
|
"epoch": 42.0,
|
|
"eval_loss": 0.8929181098937988,
|
|
"eval_runtime": 0.2812,
|
|
"eval_samples_per_second": 440.896,
|
|
"eval_steps_per_second": 14.222,
|
|
"step": 168
|
|
},
|
|
{
|
|
"epoch": 42.5,
|
|
"learning_rate": 4.57585828020176e-05,
|
|
"loss": 0.8584,
|
|
"step": 170
|
|
},
|
|
{
|
|
"epoch": 43.0,
|
|
"eval_loss": 0.8759724497795105,
|
|
"eval_runtime": 0.2817,
|
|
"eval_samples_per_second": 440.146,
|
|
"eval_steps_per_second": 14.198,
|
|
"step": 172
|
|
},
|
|
{
|
|
"epoch": 44.0,
|
|
"eval_loss": 0.8568333983421326,
|
|
"eval_runtime": 0.2812,
|
|
"eval_samples_per_second": 440.897,
|
|
"eval_steps_per_second": 14.222,
|
|
"step": 176
|
|
},
|
|
{
|
|
"epoch": 45.0,
|
|
"learning_rate": 4.517772654979023e-05,
|
|
"loss": 0.8127,
|
|
"step": 180
|
|
},
|
|
{
|
|
"epoch": 45.0,
|
|
"eval_loss": 0.8469016551971436,
|
|
"eval_runtime": 0.2806,
|
|
"eval_samples_per_second": 441.883,
|
|
"eval_steps_per_second": 14.254,
|
|
"step": 180
|
|
},
|
|
{
|
|
"epoch": 46.0,
|
|
"eval_loss": 0.8304460644721985,
|
|
"eval_runtime": 0.2817,
|
|
"eval_samples_per_second": 440.109,
|
|
"eval_steps_per_second": 14.197,
|
|
"step": 184
|
|
},
|
|
{
|
|
"epoch": 47.0,
|
|
"eval_loss": 0.8170428276062012,
|
|
"eval_runtime": 0.2841,
|
|
"eval_samples_per_second": 436.428,
|
|
"eval_steps_per_second": 14.078,
|
|
"step": 188
|
|
},
|
|
{
|
|
"epoch": 47.5,
|
|
"learning_rate": 4.4563803722282074e-05,
|
|
"loss": 0.7628,
|
|
"step": 190
|
|
},
|
|
{
|
|
"epoch": 48.0,
|
|
"eval_loss": 0.8019401431083679,
|
|
"eval_runtime": 0.2822,
|
|
"eval_samples_per_second": 439.371,
|
|
"eval_steps_per_second": 14.173,
|
|
"step": 192
|
|
},
|
|
{
|
|
"epoch": 49.0,
|
|
"eval_loss": 0.784856379032135,
|
|
"eval_runtime": 0.2816,
|
|
"eval_samples_per_second": 440.391,
|
|
"eval_steps_per_second": 14.206,
|
|
"step": 196
|
|
},
|
|
{
|
|
"epoch": 50.0,
|
|
"learning_rate": 4.391782039544238e-05,
|
|
"loss": 0.7346,
|
|
"step": 200
|
|
},
|
|
{
|
|
"epoch": 50.0,
|
|
"eval_loss": 0.7676337957382202,
|
|
"eval_runtime": 0.2855,
|
|
"eval_samples_per_second": 434.352,
|
|
"eval_steps_per_second": 14.011,
|
|
"step": 200
|
|
},
|
|
{
|
|
"epoch": 51.0,
|
|
"eval_loss": 0.7519186735153198,
|
|
"eval_runtime": 0.2811,
|
|
"eval_samples_per_second": 441.178,
|
|
"eval_steps_per_second": 14.232,
|
|
"step": 204
|
|
},
|
|
{
|
|
"epoch": 52.0,
|
|
"eval_loss": 0.7448477149009705,
|
|
"eval_runtime": 0.2811,
|
|
"eval_samples_per_second": 441.093,
|
|
"eval_steps_per_second": 14.229,
|
|
"step": 208
|
|
},
|
|
{
|
|
"epoch": 52.5,
|
|
"learning_rate": 4.324083518488151e-05,
|
|
"loss": 0.6752,
|
|
"step": 210
|
|
},
|
|
{
|
|
"epoch": 53.0,
|
|
"eval_loss": 0.7236030101776123,
|
|
"eval_runtime": 0.2811,
|
|
"eval_samples_per_second": 441.166,
|
|
"eval_steps_per_second": 14.231,
|
|
"step": 212
|
|
},
|
|
{
|
|
"epoch": 54.0,
|
|
"eval_loss": 0.7119696736335754,
|
|
"eval_runtime": 0.281,
|
|
"eval_samples_per_second": 441.247,
|
|
"eval_steps_per_second": 14.234,
|
|
"step": 216
|
|
},
|
|
{
|
|
"epoch": 55.0,
|
|
"learning_rate": 4.253395751104748e-05,
|
|
"loss": 0.6514,
|
|
"step": 220
|
|
},
|
|
{
|
|
"epoch": 55.0,
|
|
"eval_loss": 0.6985801458358765,
|
|
"eval_runtime": 0.2789,
|
|
"eval_samples_per_second": 444.558,
|
|
"eval_steps_per_second": 14.341,
|
|
"step": 220
|
|
},
|
|
{
|
|
"epoch": 56.0,
|
|
"eval_loss": 0.6845277547836304,
|
|
"eval_runtime": 0.2858,
|
|
"eval_samples_per_second": 433.821,
|
|
"eval_steps_per_second": 13.994,
|
|
"step": 224
|
|
},
|
|
{
|
|
"epoch": 57.0,
|
|
"eval_loss": 0.6741825938224792,
|
|
"eval_runtime": 0.2813,
|
|
"eval_samples_per_second": 440.836,
|
|
"eval_steps_per_second": 14.221,
|
|
"step": 228
|
|
},
|
|
{
|
|
"epoch": 57.5,
|
|
"learning_rate": 4.179834578114531e-05,
|
|
"loss": 0.5972,
|
|
"step": 230
|
|
},
|
|
{
|
|
"epoch": 58.0,
|
|
"eval_loss": 0.6572293639183044,
|
|
"eval_runtime": 0.2806,
|
|
"eval_samples_per_second": 441.9,
|
|
"eval_steps_per_second": 14.255,
|
|
"step": 232
|
|
},
|
|
{
|
|
"epoch": 59.0,
|
|
"eval_loss": 0.6450180411338806,
|
|
"eval_runtime": 0.2818,
|
|
"eval_samples_per_second": 440.012,
|
|
"eval_steps_per_second": 14.194,
|
|
"step": 236
|
|
},
|
|
{
|
|
"epoch": 60.0,
|
|
"learning_rate": 4.10352054907785e-05,
|
|
"loss": 0.5653,
|
|
"step": 240
|
|
},
|
|
{
|
|
"epoch": 60.0,
|
|
"eval_loss": 0.6229264140129089,
|
|
"eval_runtime": 0.2966,
|
|
"eval_samples_per_second": 418.055,
|
|
"eval_steps_per_second": 13.486,
|
|
"step": 240
|
|
},
|
|
{
|
|
"epoch": 61.0,
|
|
"eval_loss": 0.6052446961402893,
|
|
"eval_runtime": 0.2811,
|
|
"eval_samples_per_second": 441.187,
|
|
"eval_steps_per_second": 14.232,
|
|
"step": 244
|
|
},
|
|
{
|
|
"epoch": 62.0,
|
|
"eval_loss": 0.5894925594329834,
|
|
"eval_runtime": 0.2807,
|
|
"eval_samples_per_second": 441.8,
|
|
"eval_steps_per_second": 14.252,
|
|
"step": 248
|
|
},
|
|
{
|
|
"epoch": 62.5,
|
|
"learning_rate": 4.0245787248423614e-05,
|
|
"loss": 0.521,
|
|
"step": 250
|
|
},
|
|
{
|
|
"epoch": 63.0,
|
|
"eval_loss": 0.5734381079673767,
|
|
"eval_runtime": 0.2859,
|
|
"eval_samples_per_second": 433.719,
|
|
"eval_steps_per_second": 13.991,
|
|
"step": 252
|
|
},
|
|
{
|
|
"epoch": 64.0,
|
|
"eval_loss": 0.5632374882698059,
|
|
"eval_runtime": 0.2857,
|
|
"eval_samples_per_second": 434.093,
|
|
"eval_steps_per_second": 14.003,
|
|
"step": 256
|
|
},
|
|
{
|
|
"epoch": 65.0,
|
|
"learning_rate": 3.951391154534415e-05,
|
|
"loss": 0.5041,
|
|
"step": 260
|
|
},
|
|
{
|
|
"epoch": 65.0,
|
|
"eval_loss": 0.5504010319709778,
|
|
"eval_runtime": 0.28,
|
|
"eval_samples_per_second": 442.867,
|
|
"eval_steps_per_second": 14.286,
|
|
"step": 260
|
|
},
|
|
{
|
|
"epoch": 66.0,
|
|
"eval_loss": 0.5414384603500366,
|
|
"eval_runtime": 0.2852,
|
|
"eval_samples_per_second": 434.83,
|
|
"eval_steps_per_second": 14.027,
|
|
"step": 264
|
|
},
|
|
{
|
|
"epoch": 67.0,
|
|
"eval_loss": 0.5338427424430847,
|
|
"eval_runtime": 0.2821,
|
|
"eval_samples_per_second": 439.555,
|
|
"eval_steps_per_second": 14.179,
|
|
"step": 268
|
|
},
|
|
{
|
|
"epoch": 67.5,
|
|
"learning_rate": 3.876276956764509e-05,
|
|
"loss": 0.4637,
|
|
"step": 270
|
|
},
|
|
{
|
|
"epoch": 68.0,
|
|
"eval_loss": 0.5199601650238037,
|
|
"eval_runtime": 0.2797,
|
|
"eval_samples_per_second": 443.269,
|
|
"eval_steps_per_second": 14.299,
|
|
"step": 272
|
|
},
|
|
{
|
|
"epoch": 69.0,
|
|
"eval_loss": 0.5037544369697571,
|
|
"eval_runtime": 0.2797,
|
|
"eval_samples_per_second": 443.356,
|
|
"eval_steps_per_second": 14.302,
|
|
"step": 276
|
|
},
|
|
{
|
|
"epoch": 70.0,
|
|
"learning_rate": 3.7906784690894645e-05,
|
|
"loss": 0.4242,
|
|
"step": 280
|
|
},
|
|
{
|
|
"epoch": 70.0,
|
|
"eval_loss": 0.48893558979034424,
|
|
"eval_runtime": 0.2801,
|
|
"eval_samples_per_second": 442.758,
|
|
"eval_steps_per_second": 14.283,
|
|
"step": 280
|
|
},
|
|
{
|
|
"epoch": 71.0,
|
|
"eval_loss": 0.4763914942741394,
|
|
"eval_runtime": 0.2858,
|
|
"eval_samples_per_second": 433.889,
|
|
"eval_steps_per_second": 13.996,
|
|
"step": 284
|
|
},
|
|
{
|
|
"epoch": 72.0,
|
|
"eval_loss": 0.46161094307899475,
|
|
"eval_runtime": 0.2828,
|
|
"eval_samples_per_second": 438.474,
|
|
"eval_steps_per_second": 14.144,
|
|
"step": 288
|
|
},
|
|
{
|
|
"epoch": 72.5,
|
|
"learning_rate": 3.702964861227013e-05,
|
|
"loss": 0.4153,
|
|
"step": 290
|
|
},
|
|
{
|
|
"epoch": 73.0,
|
|
"eval_loss": 0.451792448759079,
|
|
"eval_runtime": 0.2796,
|
|
"eval_samples_per_second": 443.546,
|
|
"eval_steps_per_second": 14.308,
|
|
"step": 292
|
|
},
|
|
{
|
|
"epoch": 74.0,
|
|
"eval_loss": 0.432504266500473,
|
|
"eval_runtime": 0.2805,
|
|
"eval_samples_per_second": 442.128,
|
|
"eval_steps_per_second": 14.262,
|
|
"step": 296
|
|
},
|
|
{
|
|
"epoch": 75.0,
|
|
"learning_rate": 3.613279875268731e-05,
|
|
"loss": 0.3741,
|
|
"step": 300
|
|
},
|
|
{
|
|
"epoch": 75.0,
|
|
"eval_loss": 0.42095792293548584,
|
|
"eval_runtime": 0.284,
|
|
"eval_samples_per_second": 436.647,
|
|
"eval_steps_per_second": 14.085,
|
|
"step": 300
|
|
},
|
|
{
|
|
"epoch": 76.0,
|
|
"eval_loss": 0.40618810057640076,
|
|
"eval_runtime": 0.2802,
|
|
"eval_samples_per_second": 442.485,
|
|
"eval_steps_per_second": 14.274,
|
|
"step": 304
|
|
},
|
|
{
|
|
"epoch": 77.0,
|
|
"eval_loss": 0.39309847354888916,
|
|
"eval_runtime": 0.2808,
|
|
"eval_samples_per_second": 441.66,
|
|
"eval_steps_per_second": 14.247,
|
|
"step": 308
|
|
},
|
|
{
|
|
"epoch": 77.5,
|
|
"learning_rate": 3.521770483933891e-05,
|
|
"loss": 0.3507,
|
|
"step": 310
|
|
},
|
|
{
|
|
"epoch": 78.0,
|
|
"eval_loss": 0.38285717368125916,
|
|
"eval_runtime": 0.2803,
|
|
"eval_samples_per_second": 442.404,
|
|
"eval_steps_per_second": 14.271,
|
|
"step": 312
|
|
},
|
|
{
|
|
"epoch": 79.0,
|
|
"eval_loss": 0.37365588545799255,
|
|
"eval_runtime": 0.2803,
|
|
"eval_samples_per_second": 442.446,
|
|
"eval_steps_per_second": 14.272,
|
|
"step": 316
|
|
},
|
|
{
|
|
"epoch": 80.0,
|
|
"learning_rate": 3.4285866497155414e-05,
|
|
"loss": 0.3175,
|
|
"step": 320
|
|
},
|
|
{
|
|
"epoch": 80.0,
|
|
"eval_loss": 0.3642733693122864,
|
|
"eval_runtime": 0.2806,
|
|
"eval_samples_per_second": 441.955,
|
|
"eval_steps_per_second": 14.257,
|
|
"step": 320
|
|
},
|
|
{
|
|
"epoch": 81.0,
|
|
"eval_loss": 0.3511873781681061,
|
|
"eval_runtime": 0.2811,
|
|
"eval_samples_per_second": 441.062,
|
|
"eval_steps_per_second": 14.228,
|
|
"step": 324
|
|
},
|
|
{
|
|
"epoch": 82.0,
|
|
"eval_loss": 0.34015920758247375,
|
|
"eval_runtime": 0.2821,
|
|
"eval_samples_per_second": 439.513,
|
|
"eval_steps_per_second": 14.178,
|
|
"step": 328
|
|
},
|
|
{
|
|
"epoch": 82.5,
|
|
"learning_rate": 3.333881079127052e-05,
|
|
"loss": 0.2952,
|
|
"step": 330
|
|
},
|
|
{
|
|
"epoch": 83.0,
|
|
"eval_loss": 0.3308580219745636,
|
|
"eval_runtime": 0.2805,
|
|
"eval_samples_per_second": 442.056,
|
|
"eval_steps_per_second": 14.26,
|
|
"step": 332
|
|
},
|
|
{
|
|
"epoch": 84.0,
|
|
"eval_loss": 0.31567347049713135,
|
|
"eval_runtime": 0.2816,
|
|
"eval_samples_per_second": 440.327,
|
|
"eval_steps_per_second": 14.204,
|
|
"step": 336
|
|
},
|
|
{
|
|
"epoch": 85.0,
|
|
"learning_rate": 3.2378089724518465e-05,
|
|
"loss": 0.2691,
|
|
"step": 340
|
|
},
|
|
{
|
|
"epoch": 85.0,
|
|
"eval_loss": 0.3065517842769623,
|
|
"eval_runtime": 0.2806,
|
|
"eval_samples_per_second": 441.851,
|
|
"eval_steps_per_second": 14.253,
|
|
"step": 340
|
|
},
|
|
{
|
|
"epoch": 86.0,
|
|
"eval_loss": 0.29837048053741455,
|
|
"eval_runtime": 0.2853,
|
|
"eval_samples_per_second": 434.69,
|
|
"eval_steps_per_second": 14.022,
|
|
"step": 344
|
|
},
|
|
{
|
|
"epoch": 87.0,
|
|
"eval_loss": 0.29335495829582214,
|
|
"eval_runtime": 0.28,
|
|
"eval_samples_per_second": 442.822,
|
|
"eval_steps_per_second": 14.285,
|
|
"step": 348
|
|
},
|
|
{
|
|
"epoch": 87.5,
|
|
"learning_rate": 3.1405277694064305e-05,
|
|
"loss": 0.2509,
|
|
"step": 350
|
|
},
|
|
{
|
|
"epoch": 88.0,
|
|
"eval_loss": 0.28385087847709656,
|
|
"eval_runtime": 0.2807,
|
|
"eval_samples_per_second": 441.691,
|
|
"eval_steps_per_second": 14.248,
|
|
"step": 352
|
|
},
|
|
{
|
|
"epoch": 89.0,
|
|
"eval_loss": 0.27153801918029785,
|
|
"eval_runtime": 0.2776,
|
|
"eval_samples_per_second": 446.639,
|
|
"eval_steps_per_second": 14.408,
|
|
"step": 356
|
|
},
|
|
{
|
|
"epoch": 90.0,
|
|
"learning_rate": 3.0421968911335196e-05,
|
|
"loss": 0.2181,
|
|
"step": 360
|
|
},
|
|
{
|
|
"epoch": 90.0,
|
|
"eval_loss": 0.26443856954574585,
|
|
"eval_runtime": 0.2802,
|
|
"eval_samples_per_second": 442.5,
|
|
"eval_steps_per_second": 14.274,
|
|
"step": 360
|
|
},
|
|
{
|
|
"epoch": 91.0,
|
|
"eval_loss": 0.2611262798309326,
|
|
"eval_runtime": 0.2807,
|
|
"eval_samples_per_second": 441.746,
|
|
"eval_steps_per_second": 14.25,
|
|
"step": 364
|
|
},
|
|
{
|
|
"epoch": 92.0,
|
|
"eval_loss": 0.2542667090892792,
|
|
"eval_runtime": 0.284,
|
|
"eval_samples_per_second": 436.656,
|
|
"eval_steps_per_second": 14.086,
|
|
"step": 368
|
|
},
|
|
{
|
|
"epoch": 92.5,
|
|
"learning_rate": 2.9429774789480575e-05,
|
|
"loss": 0.1994,
|
|
"step": 370
|
|
},
|
|
{
|
|
"epoch": 93.0,
|
|
"eval_loss": 0.24470169842243195,
|
|
"eval_runtime": 0.2797,
|
|
"eval_samples_per_second": 443.299,
|
|
"eval_steps_per_second": 14.3,
|
|
"step": 372
|
|
},
|
|
{
|
|
"epoch": 94.0,
|
|
"eval_loss": 0.23626084625720978,
|
|
"eval_runtime": 0.2798,
|
|
"eval_samples_per_second": 443.209,
|
|
"eval_steps_per_second": 14.297,
|
|
"step": 376
|
|
},
|
|
{
|
|
"epoch": 95.0,
|
|
"learning_rate": 2.843032130264289e-05,
|
|
"loss": 0.1787,
|
|
"step": 380
|
|
},
|
|
{
|
|
"epoch": 95.0,
|
|
"eval_loss": 0.23020628094673157,
|
|
"eval_runtime": 0.2798,
|
|
"eval_samples_per_second": 443.155,
|
|
"eval_steps_per_second": 14.295,
|
|
"step": 380
|
|
},
|
|
{
|
|
"epoch": 96.0,
|
|
"eval_loss": 0.22590340673923492,
|
|
"eval_runtime": 0.2806,
|
|
"eval_samples_per_second": 441.857,
|
|
"eval_steps_per_second": 14.253,
|
|
"step": 384
|
|
},
|
|
{
|
|
"epoch": 97.0,
|
|
"eval_loss": 0.2229429930448532,
|
|
"eval_runtime": 0.2815,
|
|
"eval_samples_per_second": 440.504,
|
|
"eval_steps_per_second": 14.21,
|
|
"step": 388
|
|
},
|
|
{
|
|
"epoch": 97.5,
|
|
"learning_rate": 2.7525959904833952e-05,
|
|
"loss": 0.1804,
|
|
"step": 390
|
|
},
|
|
{
|
|
"epoch": 98.0,
|
|
"eval_loss": 0.21293945610523224,
|
|
"eval_runtime": 0.2795,
|
|
"eval_samples_per_second": 443.594,
|
|
"eval_steps_per_second": 14.309,
|
|
"step": 392
|
|
},
|
|
{
|
|
"epoch": 99.0,
|
|
"eval_loss": 0.2044430524110794,
|
|
"eval_runtime": 0.2803,
|
|
"eval_samples_per_second": 442.417,
|
|
"eval_steps_per_second": 14.272,
|
|
"step": 396
|
|
},
|
|
{
|
|
"epoch": 100.0,
|
|
"learning_rate": 2.6517233627571858e-05,
|
|
"loss": 0.1601,
|
|
"step": 400
|
|
},
|
|
{
|
|
"epoch": 100.0,
|
|
"eval_loss": 0.19909267127513885,
|
|
"eval_runtime": 0.2792,
|
|
"eval_samples_per_second": 444.137,
|
|
"eval_steps_per_second": 14.327,
|
|
"step": 400
|
|
},
|
|
{
|
|
"epoch": 101.0,
|
|
"eval_loss": 0.1917618215084076,
|
|
"eval_runtime": 0.2824,
|
|
"eval_samples_per_second": 439.094,
|
|
"eval_steps_per_second": 14.164,
|
|
"step": 404
|
|
},
|
|
{
|
|
"epoch": 102.0,
|
|
"eval_loss": 0.18300709128379822,
|
|
"eval_runtime": 0.2802,
|
|
"eval_samples_per_second": 442.589,
|
|
"eval_steps_per_second": 14.277,
|
|
"step": 408
|
|
},
|
|
{
|
|
"epoch": 102.5,
|
|
"learning_rate": 2.550602095919722e-05,
|
|
"loss": 0.1501,
|
|
"step": 410
|
|
},
|
|
{
|
|
"epoch": 103.0,
|
|
"eval_loss": 0.17836973071098328,
|
|
"eval_runtime": 0.2813,
|
|
"eval_samples_per_second": 440.753,
|
|
"eval_steps_per_second": 14.218,
|
|
"step": 412
|
|
},
|
|
{
|
|
"epoch": 104.0,
|
|
"eval_loss": 0.17364360392093658,
|
|
"eval_runtime": 0.2883,
|
|
"eval_samples_per_second": 430.109,
|
|
"eval_steps_per_second": 13.874,
|
|
"step": 416
|
|
},
|
|
{
|
|
"epoch": 105.0,
|
|
"learning_rate": 2.4493979040802785e-05,
|
|
"loss": 0.1329,
|
|
"step": 420
|
|
},
|
|
{
|
|
"epoch": 105.0,
|
|
"eval_loss": 0.16578342020511627,
|
|
"eval_runtime": 0.279,
|
|
"eval_samples_per_second": 444.52,
|
|
"eval_steps_per_second": 14.339,
|
|
"step": 420
|
|
},
|
|
{
|
|
"epoch": 106.0,
|
|
"eval_loss": 0.16024312376976013,
|
|
"eval_runtime": 0.2795,
|
|
"eval_samples_per_second": 443.575,
|
|
"eval_steps_per_second": 14.309,
|
|
"step": 424
|
|
},
|
|
{
|
|
"epoch": 107.0,
|
|
"eval_loss": 0.1547831892967224,
|
|
"eval_runtime": 0.2796,
|
|
"eval_samples_per_second": 443.526,
|
|
"eval_steps_per_second": 14.307,
|
|
"step": 428
|
|
},
|
|
{
|
|
"epoch": 107.5,
|
|
"learning_rate": 2.348276637242814e-05,
|
|
"loss": 0.1233,
|
|
"step": 430
|
|
},
|
|
{
|
|
"epoch": 108.0,
|
|
"eval_loss": 0.14960607886314392,
|
|
"eval_runtime": 0.2847,
|
|
"eval_samples_per_second": 435.541,
|
|
"eval_steps_per_second": 14.05,
|
|
"step": 432
|
|
},
|
|
{
|
|
"epoch": 109.0,
|
|
"eval_loss": 0.1460711807012558,
|
|
"eval_runtime": 0.2837,
|
|
"eval_samples_per_second": 437.097,
|
|
"eval_steps_per_second": 14.1,
|
|
"step": 436
|
|
},
|
|
{
|
|
"epoch": 110.0,
|
|
"learning_rate": 2.25747536786338e-05,
|
|
"loss": 0.1185,
|
|
"step": 440
|
|
},
|
|
{
|
|
"epoch": 110.0,
|
|
"eval_loss": 0.13765843212604523,
|
|
"eval_runtime": 0.282,
|
|
"eval_samples_per_second": 439.791,
|
|
"eval_steps_per_second": 14.187,
|
|
"step": 440
|
|
},
|
|
{
|
|
"epoch": 111.0,
|
|
"eval_loss": 0.1353219449520111,
|
|
"eval_runtime": 0.2851,
|
|
"eval_samples_per_second": 435.008,
|
|
"eval_steps_per_second": 14.033,
|
|
"step": 444
|
|
},
|
|
{
|
|
"epoch": 112.0,
|
|
"eval_loss": 0.1338249146938324,
|
|
"eval_runtime": 0.2858,
|
|
"eval_samples_per_second": 433.886,
|
|
"eval_steps_per_second": 13.996,
|
|
"step": 448
|
|
},
|
|
{
|
|
"epoch": 112.5,
|
|
"learning_rate": 2.1569678697357127e-05,
|
|
"loss": 0.1045,
|
|
"step": 450
|
|
},
|
|
{
|
|
"epoch": 113.0,
|
|
"eval_loss": 0.13033108413219452,
|
|
"eval_runtime": 0.2825,
|
|
"eval_samples_per_second": 438.977,
|
|
"eval_steps_per_second": 14.161,
|
|
"step": 452
|
|
},
|
|
{
|
|
"epoch": 114.0,
|
|
"eval_loss": 0.127507284283638,
|
|
"eval_runtime": 0.2822,
|
|
"eval_samples_per_second": 439.429,
|
|
"eval_steps_per_second": 14.175,
|
|
"step": 456
|
|
},
|
|
{
|
|
"epoch": 115.0,
|
|
"learning_rate": 2.0570225210519434e-05,
|
|
"loss": 0.0992,
|
|
"step": 460
|
|
},
|
|
{
|
|
"epoch": 115.0,
|
|
"eval_loss": 0.12555664777755737,
|
|
"eval_runtime": 0.2809,
|
|
"eval_samples_per_second": 441.479,
|
|
"eval_steps_per_second": 14.241,
|
|
"step": 460
|
|
},
|
|
{
|
|
"epoch": 116.0,
|
|
"eval_loss": 0.12143656611442566,
|
|
"eval_runtime": 0.2879,
|
|
"eval_samples_per_second": 430.769,
|
|
"eval_steps_per_second": 13.896,
|
|
"step": 464
|
|
},
|
|
{
|
|
"epoch": 117.0,
|
|
"eval_loss": 0.1160622239112854,
|
|
"eval_runtime": 0.2846,
|
|
"eval_samples_per_second": 435.628,
|
|
"eval_steps_per_second": 14.053,
|
|
"step": 468
|
|
},
|
|
{
|
|
"epoch": 117.5,
|
|
"learning_rate": 1.957803108866481e-05,
|
|
"loss": 0.0953,
|
|
"step": 470
|
|
},
|
|
{
|
|
"epoch": 118.0,
|
|
"eval_loss": 0.10977106541395187,
|
|
"eval_runtime": 0.2808,
|
|
"eval_samples_per_second": 441.636,
|
|
"eval_steps_per_second": 14.246,
|
|
"step": 472
|
|
},
|
|
{
|
|
"epoch": 119.0,
|
|
"eval_loss": 0.10678129643201828,
|
|
"eval_runtime": 0.2818,
|
|
"eval_samples_per_second": 440.075,
|
|
"eval_steps_per_second": 14.196,
|
|
"step": 476
|
|
},
|
|
{
|
|
"epoch": 120.0,
|
|
"learning_rate": 1.859472230593569e-05,
|
|
"loss": 0.0855,
|
|
"step": 480
|
|
},
|
|
{
|
|
"epoch": 120.0,
|
|
"eval_loss": 0.10289318859577179,
|
|
"eval_runtime": 0.2816,
|
|
"eval_samples_per_second": 440.278,
|
|
"eval_steps_per_second": 14.203,
|
|
"step": 480
|
|
},
|
|
{
|
|
"epoch": 121.0,
|
|
"eval_loss": 0.09968823939561844,
|
|
"eval_runtime": 0.2802,
|
|
"eval_samples_per_second": 442.616,
|
|
"eval_steps_per_second": 14.278,
|
|
"step": 484
|
|
},
|
|
{
|
|
"epoch": 122.0,
|
|
"eval_loss": 0.0967448428273201,
|
|
"eval_runtime": 0.2802,
|
|
"eval_samples_per_second": 442.498,
|
|
"eval_steps_per_second": 14.274,
|
|
"step": 488
|
|
},
|
|
{
|
|
"epoch": 122.5,
|
|
"learning_rate": 1.7621910275481544e-05,
|
|
"loss": 0.0752,
|
|
"step": 490
|
|
},
|
|
{
|
|
"epoch": 123.0,
|
|
"eval_loss": 0.09666424989700317,
|
|
"eval_runtime": 0.2802,
|
|
"eval_samples_per_second": 442.554,
|
|
"eval_steps_per_second": 14.276,
|
|
"step": 492
|
|
},
|
|
{
|
|
"epoch": 124.0,
|
|
"eval_loss": 0.09402679651975632,
|
|
"eval_runtime": 0.2825,
|
|
"eval_samples_per_second": 438.982,
|
|
"eval_steps_per_second": 14.161,
|
|
"step": 496
|
|
},
|
|
{
|
|
"epoch": 125.0,
|
|
"learning_rate": 1.685229026768593e-05,
|
|
"loss": 0.0705,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 125.0,
|
|
"eval_loss": 0.09048105031251907,
|
|
"eval_runtime": 0.2968,
|
|
"eval_samples_per_second": 417.789,
|
|
"eval_steps_per_second": 13.477,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 126.0,
|
|
"eval_loss": 0.08732099831104279,
|
|
"eval_runtime": 0.2828,
|
|
"eval_samples_per_second": 438.481,
|
|
"eval_steps_per_second": 14.145,
|
|
"step": 504
|
|
},
|
|
{
|
|
"epoch": 127.0,
|
|
"eval_loss": 0.0858684554696083,
|
|
"eval_runtime": 0.2808,
|
|
"eval_samples_per_second": 441.647,
|
|
"eval_steps_per_second": 14.247,
|
|
"step": 508
|
|
},
|
|
{
|
|
"epoch": 127.5,
|
|
"learning_rate": 1.5902376575912815e-05,
|
|
"loss": 0.066,
|
|
"step": 510
|
|
},
|
|
{
|
|
"epoch": 128.0,
|
|
"eval_loss": 0.08412772417068481,
|
|
"eval_runtime": 0.3261,
|
|
"eval_samples_per_second": 380.248,
|
|
"eval_steps_per_second": 12.266,
|
|
"step": 512
|
|
},
|
|
{
|
|
"epoch": 129.0,
|
|
"eval_loss": 0.08159559965133667,
|
|
"eval_runtime": 0.2813,
|
|
"eval_samples_per_second": 440.888,
|
|
"eval_steps_per_second": 14.222,
|
|
"step": 516
|
|
},
|
|
{
|
|
"epoch": 130.0,
|
|
"learning_rate": 1.4967371761464738e-05,
|
|
"loss": 0.062,
|
|
"step": 520
|
|
},
|
|
{
|
|
"epoch": 130.0,
|
|
"eval_loss": 0.07958081364631653,
|
|
"eval_runtime": 0.2798,
|
|
"eval_samples_per_second": 443.168,
|
|
"eval_steps_per_second": 14.296,
|
|
"step": 520
|
|
},
|
|
{
|
|
"epoch": 131.0,
|
|
"eval_loss": 0.07724415510892868,
|
|
"eval_runtime": 0.28,
|
|
"eval_samples_per_second": 442.83,
|
|
"eval_steps_per_second": 14.285,
|
|
"step": 524
|
|
},
|
|
{
|
|
"epoch": 132.0,
|
|
"eval_loss": 0.07607536762952805,
|
|
"eval_runtime": 0.2805,
|
|
"eval_samples_per_second": 442.123,
|
|
"eval_steps_per_second": 14.262,
|
|
"step": 528
|
|
},
|
|
{
|
|
"epoch": 132.5,
|
|
"learning_rate": 1.4048808078582942e-05,
|
|
"loss": 0.0546,
|
|
"step": 530
|
|
},
|
|
{
|
|
"epoch": 133.0,
|
|
"eval_loss": 0.07420383393764496,
|
|
"eval_runtime": 0.2793,
|
|
"eval_samples_per_second": 443.905,
|
|
"eval_steps_per_second": 14.32,
|
|
"step": 532
|
|
},
|
|
{
|
|
"epoch": 134.0,
|
|
"eval_loss": 0.07149375230073929,
|
|
"eval_runtime": 0.28,
|
|
"eval_samples_per_second": 442.783,
|
|
"eval_steps_per_second": 14.283,
|
|
"step": 536
|
|
},
|
|
{
|
|
"epoch": 135.0,
|
|
"learning_rate": 1.3148190838338803e-05,
|
|
"loss": 0.0574,
|
|
"step": 540
|
|
},
|
|
{
|
|
"epoch": 135.0,
|
|
"eval_loss": 0.06859742105007172,
|
|
"eval_runtime": 0.279,
|
|
"eval_samples_per_second": 444.412,
|
|
"eval_steps_per_second": 14.336,
|
|
"step": 540
|
|
},
|
|
{
|
|
"epoch": 136.0,
|
|
"eval_loss": 0.06600970029830933,
|
|
"eval_runtime": 0.28,
|
|
"eval_samples_per_second": 442.837,
|
|
"eval_steps_per_second": 14.285,
|
|
"step": 544
|
|
},
|
|
{
|
|
"epoch": 137.0,
|
|
"eval_loss": 0.06500796228647232,
|
|
"eval_runtime": 0.281,
|
|
"eval_samples_per_second": 441.342,
|
|
"eval_steps_per_second": 14.237,
|
|
"step": 548
|
|
},
|
|
{
|
|
"epoch": 137.5,
|
|
"learning_rate": 1.2266995941780934e-05,
|
|
"loss": 0.0495,
|
|
"step": 550
|
|
},
|
|
{
|
|
"epoch": 138.0,
|
|
"eval_loss": 0.06258592754602432,
|
|
"eval_runtime": 0.2813,
|
|
"eval_samples_per_second": 440.754,
|
|
"eval_steps_per_second": 14.218,
|
|
"step": 552
|
|
},
|
|
{
|
|
"epoch": 139.0,
|
|
"eval_loss": 0.06007996201515198,
|
|
"eval_runtime": 0.282,
|
|
"eval_samples_per_second": 439.742,
|
|
"eval_steps_per_second": 14.185,
|
|
"step": 556
|
|
},
|
|
{
|
|
"epoch": 140.0,
|
|
"learning_rate": 1.140666746127854e-05,
|
|
"loss": 0.0469,
|
|
"step": 560
|
|
},
|
|
{
|
|
"epoch": 140.0,
|
|
"eval_loss": 0.05839233472943306,
|
|
"eval_runtime": 0.2805,
|
|
"eval_samples_per_second": 442.1,
|
|
"eval_steps_per_second": 14.261,
|
|
"step": 560
|
|
},
|
|
{
|
|
"epoch": 141.0,
|
|
"eval_loss": 0.05694788321852684,
|
|
"eval_runtime": 0.2798,
|
|
"eval_samples_per_second": 443.162,
|
|
"eval_steps_per_second": 14.296,
|
|
"step": 564
|
|
},
|
|
{
|
|
"epoch": 142.0,
|
|
"eval_loss": 0.055415015667676926,
|
|
"eval_runtime": 0.2857,
|
|
"eval_samples_per_second": 433.955,
|
|
"eval_steps_per_second": 13.999,
|
|
"step": 568
|
|
},
|
|
{
|
|
"epoch": 142.5,
|
|
"learning_rate": 1.0568615274024522e-05,
|
|
"loss": 0.0447,
|
|
"step": 570
|
|
},
|
|
{
|
|
"epoch": 143.0,
|
|
"eval_loss": 0.055186230689287186,
|
|
"eval_runtime": 0.28,
|
|
"eval_samples_per_second": 442.93,
|
|
"eval_steps_per_second": 14.288,
|
|
"step": 572
|
|
},
|
|
{
|
|
"epoch": 144.0,
|
|
"eval_loss": 0.05495119094848633,
|
|
"eval_runtime": 0.2906,
|
|
"eval_samples_per_second": 426.752,
|
|
"eval_steps_per_second": 13.766,
|
|
"step": 576
|
|
},
|
|
{
|
|
"epoch": 145.0,
|
|
"learning_rate": 9.754212751576386e-06,
|
|
"loss": 0.0425,
|
|
"step": 580
|
|
},
|
|
{
|
|
"epoch": 145.0,
|
|
"eval_loss": 0.053993016481399536,
|
|
"eval_runtime": 0.2807,
|
|
"eval_samples_per_second": 441.809,
|
|
"eval_steps_per_second": 14.252,
|
|
"step": 580
|
|
},
|
|
{
|
|
"epoch": 146.0,
|
|
"eval_loss": 0.052898697555065155,
|
|
"eval_runtime": 0.2799,
|
|
"eval_samples_per_second": 442.978,
|
|
"eval_steps_per_second": 14.29,
|
|
"step": 584
|
|
},
|
|
{
|
|
"epoch": 147.0,
|
|
"eval_loss": 0.051423486322164536,
|
|
"eval_runtime": 0.2829,
|
|
"eval_samples_per_second": 438.379,
|
|
"eval_steps_per_second": 14.141,
|
|
"step": 588
|
|
},
|
|
{
|
|
"epoch": 147.5,
|
|
"learning_rate": 8.964794509221508e-06,
|
|
"loss": 0.0389,
|
|
"step": 590
|
|
},
|
|
{
|
|
"epoch": 148.0,
|
|
"eval_loss": 0.049843259155750275,
|
|
"eval_runtime": 0.2804,
|
|
"eval_samples_per_second": 442.265,
|
|
"eval_steps_per_second": 14.267,
|
|
"step": 592
|
|
},
|
|
{
|
|
"epoch": 149.0,
|
|
"eval_loss": 0.04916330799460411,
|
|
"eval_runtime": 0.2794,
|
|
"eval_samples_per_second": 443.845,
|
|
"eval_steps_per_second": 14.318,
|
|
"step": 596
|
|
},
|
|
{
|
|
"epoch": 150.0,
|
|
"learning_rate": 8.20165421885469e-06,
|
|
"loss": 0.0398,
|
|
"step": 600
|
|
},
|
|
{
|
|
"epoch": 150.0,
|
|
"eval_loss": 0.04869178682565689,
|
|
"eval_runtime": 0.287,
|
|
"eval_samples_per_second": 432.016,
|
|
"eval_steps_per_second": 13.936,
|
|
"step": 600
|
|
},
|
|
{
|
|
"epoch": 151.0,
|
|
"eval_loss": 0.047069448977708817,
|
|
"eval_runtime": 0.2817,
|
|
"eval_samples_per_second": 440.136,
|
|
"eval_steps_per_second": 14.198,
|
|
"step": 604
|
|
},
|
|
{
|
|
"epoch": 152.0,
|
|
"eval_loss": 0.04603598266839981,
|
|
"eval_runtime": 0.2867,
|
|
"eval_samples_per_second": 432.555,
|
|
"eval_steps_per_second": 13.953,
|
|
"step": 608
|
|
},
|
|
{
|
|
"epoch": 152.5,
|
|
"learning_rate": 7.466042488952521e-06,
|
|
"loss": 0.0371,
|
|
"step": 610
|
|
},
|
|
{
|
|
"epoch": 153.0,
|
|
"eval_loss": 0.04540390521287918,
|
|
"eval_runtime": 0.2793,
|
|
"eval_samples_per_second": 443.918,
|
|
"eval_steps_per_second": 14.32,
|
|
"step": 612
|
|
},
|
|
{
|
|
"epoch": 154.0,
|
|
"eval_loss": 0.04474242404103279,
|
|
"eval_runtime": 0.2897,
|
|
"eval_samples_per_second": 427.957,
|
|
"eval_steps_per_second": 13.805,
|
|
"step": 616
|
|
},
|
|
{
|
|
"epoch": 155.0,
|
|
"learning_rate": 6.7591648151184935e-06,
|
|
"loss": 0.0338,
|
|
"step": 620
|
|
},
|
|
{
|
|
"epoch": 155.0,
|
|
"eval_loss": 0.04416840150952339,
|
|
"eval_runtime": 0.2799,
|
|
"eval_samples_per_second": 443.027,
|
|
"eval_steps_per_second": 14.291,
|
|
"step": 620
|
|
},
|
|
{
|
|
"epoch": 156.0,
|
|
"eval_loss": 0.043181490153074265,
|
|
"eval_runtime": 0.2822,
|
|
"eval_samples_per_second": 439.481,
|
|
"eval_steps_per_second": 14.177,
|
|
"step": 624
|
|
},
|
|
{
|
|
"epoch": 157.0,
|
|
"eval_loss": 0.04268254339694977,
|
|
"eval_runtime": 0.2795,
|
|
"eval_samples_per_second": 443.645,
|
|
"eval_steps_per_second": 14.311,
|
|
"step": 628
|
|
},
|
|
{
|
|
"epoch": 157.5,
|
|
"learning_rate": 6.148500953954992e-06,
|
|
"loss": 0.0333,
|
|
"step": 630
|
|
},
|
|
{
|
|
"epoch": 158.0,
|
|
"eval_loss": 0.04142157360911369,
|
|
"eval_runtime": 0.2796,
|
|
"eval_samples_per_second": 443.495,
|
|
"eval_steps_per_second": 14.306,
|
|
"step": 632
|
|
},
|
|
{
|
|
"epoch": 159.0,
|
|
"eval_loss": 0.040608469396829605,
|
|
"eval_runtime": 0.2788,
|
|
"eval_samples_per_second": 444.766,
|
|
"eval_steps_per_second": 14.347,
|
|
"step": 636
|
|
},
|
|
{
|
|
"epoch": 160.0,
|
|
"learning_rate": 5.499368948885528e-06,
|
|
"loss": 0.0329,
|
|
"step": 640
|
|
},
|
|
{
|
|
"epoch": 160.0,
|
|
"eval_loss": 0.039851173758506775,
|
|
"eval_runtime": 0.2803,
|
|
"eval_samples_per_second": 442.396,
|
|
"eval_steps_per_second": 14.271,
|
|
"step": 640
|
|
},
|
|
{
|
|
"epoch": 161.0,
|
|
"eval_loss": 0.03928506746888161,
|
|
"eval_runtime": 0.2802,
|
|
"eval_samples_per_second": 442.597,
|
|
"eval_steps_per_second": 14.277,
|
|
"step": 644
|
|
},
|
|
{
|
|
"epoch": 162.0,
|
|
"eval_loss": 0.038857731968164444,
|
|
"eval_runtime": 0.2813,
|
|
"eval_samples_per_second": 440.849,
|
|
"eval_steps_per_second": 14.221,
|
|
"step": 648
|
|
},
|
|
{
|
|
"epoch": 162.5,
|
|
"learning_rate": 4.882193917912398e-06,
|
|
"loss": 0.0308,
|
|
"step": 650
|
|
},
|
|
{
|
|
"epoch": 163.0,
|
|
"eval_loss": 0.0385042242705822,
|
|
"eval_runtime": 0.2819,
|
|
"eval_samples_per_second": 439.805,
|
|
"eval_steps_per_second": 14.187,
|
|
"step": 652
|
|
},
|
|
{
|
|
"epoch": 164.0,
|
|
"eval_loss": 0.03806820511817932,
|
|
"eval_runtime": 0.2822,
|
|
"eval_samples_per_second": 439.449,
|
|
"eval_steps_per_second": 14.176,
|
|
"step": 656
|
|
},
|
|
{
|
|
"epoch": 165.0,
|
|
"learning_rate": 4.297987266585826e-06,
|
|
"loss": 0.0305,
|
|
"step": 660
|
|
},
|
|
{
|
|
"epoch": 165.0,
|
|
"eval_loss": 0.037549860775470734,
|
|
"eval_runtime": 0.2826,
|
|
"eval_samples_per_second": 438.719,
|
|
"eval_steps_per_second": 14.152,
|
|
"step": 660
|
|
},
|
|
{
|
|
"epoch": 166.0,
|
|
"eval_loss": 0.037020985037088394,
|
|
"eval_runtime": 0.2801,
|
|
"eval_samples_per_second": 442.7,
|
|
"eval_steps_per_second": 14.281,
|
|
"step": 664
|
|
},
|
|
{
|
|
"epoch": 167.0,
|
|
"eval_loss": 0.03641499951481819,
|
|
"eval_runtime": 0.2813,
|
|
"eval_samples_per_second": 440.848,
|
|
"eval_steps_per_second": 14.221,
|
|
"step": 668
|
|
},
|
|
{
|
|
"epoch": 167.5,
|
|
"learning_rate": 3.7477063729913804e-06,
|
|
"loss": 0.0311,
|
|
"step": 670
|
|
},
|
|
{
|
|
"epoch": 168.0,
|
|
"eval_loss": 0.035942208021879196,
|
|
"eval_runtime": 0.281,
|
|
"eval_samples_per_second": 441.343,
|
|
"eval_steps_per_second": 14.237,
|
|
"step": 672
|
|
},
|
|
{
|
|
"epoch": 169.0,
|
|
"eval_loss": 0.03532649576663971,
|
|
"eval_runtime": 0.2806,
|
|
"eval_samples_per_second": 441.977,
|
|
"eval_steps_per_second": 14.257,
|
|
"step": 676
|
|
},
|
|
{
|
|
"epoch": 170.0,
|
|
"learning_rate": 3.232253018831208e-06,
|
|
"loss": 0.0296,
|
|
"step": 680
|
|
},
|
|
{
|
|
"epoch": 170.0,
|
|
"eval_loss": 0.03475267067551613,
|
|
"eval_runtime": 0.2792,
|
|
"eval_samples_per_second": 444.206,
|
|
"eval_steps_per_second": 14.329,
|
|
"step": 680
|
|
},
|
|
{
|
|
"epoch": 171.0,
|
|
"eval_loss": 0.03437618166208267,
|
|
"eval_runtime": 0.28,
|
|
"eval_samples_per_second": 442.93,
|
|
"eval_steps_per_second": 14.288,
|
|
"step": 684
|
|
},
|
|
{
|
|
"epoch": 172.0,
|
|
"eval_loss": 0.03419204428792,
|
|
"eval_runtime": 0.2813,
|
|
"eval_samples_per_second": 440.794,
|
|
"eval_steps_per_second": 14.219,
|
|
"step": 688
|
|
},
|
|
{
|
|
"epoch": 172.5,
|
|
"learning_rate": 2.7524719116147153e-06,
|
|
"loss": 0.0291,
|
|
"step": 690
|
|
},
|
|
{
|
|
"epoch": 173.0,
|
|
"eval_loss": 0.03385160118341446,
|
|
"eval_runtime": 0.2802,
|
|
"eval_samples_per_second": 442.519,
|
|
"eval_steps_per_second": 14.275,
|
|
"step": 692
|
|
},
|
|
{
|
|
"epoch": 174.0,
|
|
"eval_loss": 0.03362323343753815,
|
|
"eval_runtime": 0.2798,
|
|
"eval_samples_per_second": 443.163,
|
|
"eval_steps_per_second": 14.296,
|
|
"step": 696
|
|
},
|
|
{
|
|
"epoch": 175.0,
|
|
"learning_rate": 2.3091493003804476e-06,
|
|
"loss": 0.0282,
|
|
"step": 700
|
|
},
|
|
{
|
|
"epoch": 175.0,
|
|
"eval_loss": 0.03348943591117859,
|
|
"eval_runtime": 0.2796,
|
|
"eval_samples_per_second": 443.553,
|
|
"eval_steps_per_second": 14.308,
|
|
"step": 700
|
|
},
|
|
{
|
|
"epoch": 176.0,
|
|
"eval_loss": 0.033210765570402145,
|
|
"eval_runtime": 0.2798,
|
|
"eval_samples_per_second": 443.192,
|
|
"eval_steps_per_second": 14.297,
|
|
"step": 704
|
|
},
|
|
{
|
|
"epoch": 177.0,
|
|
"eval_loss": 0.03299249708652496,
|
|
"eval_runtime": 0.294,
|
|
"eval_samples_per_second": 421.822,
|
|
"eval_steps_per_second": 13.607,
|
|
"step": 708
|
|
},
|
|
{
|
|
"epoch": 177.5,
|
|
"learning_rate": 1.9030116872178316e-06,
|
|
"loss": 0.0269,
|
|
"step": 710
|
|
},
|
|
{
|
|
"epoch": 178.0,
|
|
"eval_loss": 0.03290760889649391,
|
|
"eval_runtime": 0.2855,
|
|
"eval_samples_per_second": 434.333,
|
|
"eval_steps_per_second": 14.011,
|
|
"step": 712
|
|
},
|
|
{
|
|
"epoch": 179.0,
|
|
"eval_loss": 0.03291478008031845,
|
|
"eval_runtime": 0.2795,
|
|
"eval_samples_per_second": 443.713,
|
|
"eval_steps_per_second": 14.313,
|
|
"step": 716
|
|
},
|
|
{
|
|
"epoch": 180.0,
|
|
"learning_rate": 1.5347246367000995e-06,
|
|
"loss": 0.0267,
|
|
"step": 720
|
|
},
|
|
{
|
|
"epoch": 180.0,
|
|
"eval_loss": 0.032794609665870667,
|
|
"eval_runtime": 0.2807,
|
|
"eval_samples_per_second": 441.756,
|
|
"eval_steps_per_second": 14.25,
|
|
"step": 720
|
|
},
|
|
{
|
|
"epoch": 181.0,
|
|
"eval_loss": 0.03267654404044151,
|
|
"eval_runtime": 0.2807,
|
|
"eval_samples_per_second": 441.815,
|
|
"eval_steps_per_second": 14.252,
|
|
"step": 724
|
|
},
|
|
{
|
|
"epoch": 182.0,
|
|
"eval_loss": 0.03255731984972954,
|
|
"eval_runtime": 0.2808,
|
|
"eval_samples_per_second": 441.631,
|
|
"eval_steps_per_second": 14.246,
|
|
"step": 728
|
|
},
|
|
{
|
|
"epoch": 182.5,
|
|
"learning_rate": 1.2048916851796099e-06,
|
|
"loss": 0.027,
|
|
"step": 730
|
|
},
|
|
{
|
|
"epoch": 183.0,
|
|
"eval_loss": 0.03246257081627846,
|
|
"eval_runtime": 0.2807,
|
|
"eval_samples_per_second": 441.772,
|
|
"eval_steps_per_second": 14.251,
|
|
"step": 732
|
|
},
|
|
{
|
|
"epoch": 184.0,
|
|
"eval_loss": 0.03235262259840965,
|
|
"eval_runtime": 0.2796,
|
|
"eval_samples_per_second": 443.483,
|
|
"eval_steps_per_second": 14.306,
|
|
"step": 736
|
|
},
|
|
{
|
|
"epoch": 185.0,
|
|
"learning_rate": 9.140533517329213e-07,
|
|
"loss": 0.0277,
|
|
"step": 740
|
|
},
|
|
{
|
|
"epoch": 185.0,
|
|
"eval_loss": 0.03228095546364784,
|
|
"eval_runtime": 0.2788,
|
|
"eval_samples_per_second": 444.807,
|
|
"eval_steps_per_second": 14.349,
|
|
"step": 740
|
|
},
|
|
{
|
|
"epoch": 186.0,
|
|
"eval_loss": 0.032180700451135635,
|
|
"eval_runtime": 0.2809,
|
|
"eval_samples_per_second": 441.446,
|
|
"eval_steps_per_second": 14.24,
|
|
"step": 744
|
|
},
|
|
{
|
|
"epoch": 187.0,
|
|
"eval_loss": 0.03214586153626442,
|
|
"eval_runtime": 0.2803,
|
|
"eval_samples_per_second": 442.366,
|
|
"eval_steps_per_second": 14.27,
|
|
"step": 748
|
|
},
|
|
{
|
|
"epoch": 187.5,
|
|
"learning_rate": 6.626862523763904e-07,
|
|
"loss": 0.0265,
|
|
"step": 750
|
|
},
|
|
{
|
|
"epoch": 188.0,
|
|
"eval_loss": 0.03211435675621033,
|
|
"eval_runtime": 0.2806,
|
|
"eval_samples_per_second": 441.875,
|
|
"eval_steps_per_second": 14.254,
|
|
"step": 752
|
|
},
|
|
{
|
|
"epoch": 189.0,
|
|
"eval_loss": 0.0320197194814682,
|
|
"eval_runtime": 0.281,
|
|
"eval_samples_per_second": 441.204,
|
|
"eval_steps_per_second": 14.232,
|
|
"step": 756
|
|
},
|
|
{
|
|
"epoch": 190.0,
|
|
"learning_rate": 4.512023190040071e-07,
|
|
"loss": 0.0254,
|
|
"step": 760
|
|
},
|
|
{
|
|
"epoch": 190.0,
|
|
"eval_loss": 0.03190222755074501,
|
|
"eval_runtime": 0.2801,
|
|
"eval_samples_per_second": 442.709,
|
|
"eval_steps_per_second": 14.281,
|
|
"step": 760
|
|
},
|
|
{
|
|
"epoch": 191.0,
|
|
"eval_loss": 0.03179073706269264,
|
|
"eval_runtime": 0.2812,
|
|
"eval_samples_per_second": 441.03,
|
|
"eval_steps_per_second": 14.227,
|
|
"step": 764
|
|
},
|
|
{
|
|
"epoch": 192.0,
|
|
"eval_loss": 0.031745798885822296,
|
|
"eval_runtime": 0.2809,
|
|
"eval_samples_per_second": 441.477,
|
|
"eval_steps_per_second": 14.241,
|
|
"step": 768
|
|
},
|
|
{
|
|
"epoch": 192.5,
|
|
"learning_rate": 2.799481243273366e-07,
|
|
"loss": 0.0253,
|
|
"step": 770
|
|
},
|
|
{
|
|
"epoch": 193.0,
|
|
"eval_loss": 0.031719356775283813,
|
|
"eval_runtime": 0.2811,
|
|
"eval_samples_per_second": 441.179,
|
|
"eval_steps_per_second": 14.232,
|
|
"step": 772
|
|
},
|
|
{
|
|
"epoch": 194.0,
|
|
"eval_loss": 0.031690530478954315,
|
|
"eval_runtime": 0.2822,
|
|
"eval_samples_per_second": 439.367,
|
|
"eval_steps_per_second": 14.173,
|
|
"step": 776
|
|
},
|
|
{
|
|
"epoch": 195.0,
|
|
"learning_rate": 1.4920431392388413e-07,
|
|
"loss": 0.0264,
|
|
"step": 780
|
|
},
|
|
{
|
|
"epoch": 195.0,
|
|
"eval_loss": 0.031662747263908386,
|
|
"eval_runtime": 0.2867,
|
|
"eval_samples_per_second": 432.508,
|
|
"eval_steps_per_second": 13.952,
|
|
"step": 780
|
|
},
|
|
{
|
|
"epoch": 196.0,
|
|
"eval_loss": 0.031658921390771866,
|
|
"eval_runtime": 0.2804,
|
|
"eval_samples_per_second": 442.197,
|
|
"eval_steps_per_second": 14.264,
|
|
"step": 784
|
|
},
|
|
{
|
|
"epoch": 197.0,
|
|
"eval_loss": 0.03164348006248474,
|
|
"eval_runtime": 0.2803,
|
|
"eval_samples_per_second": 442.351,
|
|
"eval_steps_per_second": 14.269,
|
|
"step": 788
|
|
},
|
|
{
|
|
"epoch": 197.5,
|
|
"learning_rate": 5.9185146324639006e-08,
|
|
"loss": 0.0263,
|
|
"step": 790
|
|
},
|
|
{
|
|
"epoch": 198.0,
|
|
"eval_loss": 0.031625885516405106,
|
|
"eval_runtime": 0.2803,
|
|
"eval_samples_per_second": 442.349,
|
|
"eval_steps_per_second": 14.269,
|
|
"step": 792
|
|
},
|
|
{
|
|
"epoch": 199.0,
|
|
"eval_loss": 0.03165370225906372,
|
|
"eval_runtime": 0.2807,
|
|
"eval_samples_per_second": 441.715,
|
|
"eval_steps_per_second": 14.249,
|
|
"step": 796
|
|
},
|
|
{
|
|
"epoch": 200.0,
|
|
"learning_rate": 1.0038141894436192e-08,
|
|
"loss": 0.0264,
|
|
"step": 800
|
|
},
|
|
{
|
|
"epoch": 200.0,
|
|
"eval_loss": 0.03163384273648262,
|
|
"eval_runtime": 0.2807,
|
|
"eval_samples_per_second": 441.7,
|
|
"eval_steps_per_second": 14.248,
|
|
"step": 800
|
|
},
|
|
{
|
|
"epoch": 200.0,
|
|
"step": 800,
|
|
"total_flos": 5.1777466223139226e+17,
|
|
"train_loss": 0.4552628890797496,
|
|
"train_runtime": 604.0745,
|
|
"train_samples_per_second": 163.887,
|
|
"train_steps_per_second": 1.324
|
|
}
|
|
],
|
|
"max_steps": 800,
|
|
"num_train_epochs": 200,
|
|
"total_flos": 5.1777466223139226e+17,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|