|
{
|
|
"best_metric": 0.8260869565217391,
|
|
"best_model_checkpoint": "swiftformer-xs-ve-U13-b-80d\\checkpoint-117",
|
|
"epoch": 64.61538461538461,
|
|
"eval_steps": 500,
|
|
"global_step": 420,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"epoch": 0.92,
|
|
"eval_accuracy": 0.21739130434782608,
|
|
"eval_loss": 1.38477623462677,
|
|
"eval_runtime": 0.7472,
|
|
"eval_samples_per_second": 61.564,
|
|
"eval_steps_per_second": 2.677,
|
|
"step": 6
|
|
},
|
|
{
|
|
"epoch": 1.54,
|
|
"learning_rate": 0.00014285714285714284,
|
|
"loss": 1.3838,
|
|
"step": 10
|
|
},
|
|
{
|
|
"epoch": 2.0,
|
|
"eval_accuracy": 0.1956521739130435,
|
|
"eval_loss": 1.3723431825637817,
|
|
"eval_runtime": 0.636,
|
|
"eval_samples_per_second": 72.332,
|
|
"eval_steps_per_second": 3.145,
|
|
"step": 13
|
|
},
|
|
{
|
|
"epoch": 2.92,
|
|
"eval_accuracy": 0.17391304347826086,
|
|
"eval_loss": 1.3539512157440186,
|
|
"eval_runtime": 0.6424,
|
|
"eval_samples_per_second": 71.606,
|
|
"eval_steps_per_second": 3.113,
|
|
"step": 19
|
|
},
|
|
{
|
|
"epoch": 3.08,
|
|
"learning_rate": 0.0002857142857142857,
|
|
"loss": 1.3023,
|
|
"step": 20
|
|
},
|
|
{
|
|
"epoch": 4.0,
|
|
"eval_accuracy": 0.2391304347826087,
|
|
"eval_loss": 1.3326826095581055,
|
|
"eval_runtime": 0.6389,
|
|
"eval_samples_per_second": 71.999,
|
|
"eval_steps_per_second": 3.13,
|
|
"step": 26
|
|
},
|
|
{
|
|
"epoch": 4.62,
|
|
"learning_rate": 0.0002932330827067669,
|
|
"loss": 1.1398,
|
|
"step": 30
|
|
},
|
|
{
|
|
"epoch": 4.92,
|
|
"eval_accuracy": 0.2391304347826087,
|
|
"eval_loss": 1.255542516708374,
|
|
"eval_runtime": 0.6222,
|
|
"eval_samples_per_second": 73.933,
|
|
"eval_steps_per_second": 3.214,
|
|
"step": 32
|
|
},
|
|
{
|
|
"epoch": 6.0,
|
|
"eval_accuracy": 0.391304347826087,
|
|
"eval_loss": 1.3010176420211792,
|
|
"eval_runtime": 0.6566,
|
|
"eval_samples_per_second": 70.062,
|
|
"eval_steps_per_second": 3.046,
|
|
"step": 39
|
|
},
|
|
{
|
|
"epoch": 6.15,
|
|
"learning_rate": 0.0002857142857142857,
|
|
"loss": 1.0076,
|
|
"step": 40
|
|
},
|
|
{
|
|
"epoch": 6.92,
|
|
"eval_accuracy": 0.5,
|
|
"eval_loss": 1.1956894397735596,
|
|
"eval_runtime": 0.6475,
|
|
"eval_samples_per_second": 71.037,
|
|
"eval_steps_per_second": 3.089,
|
|
"step": 45
|
|
},
|
|
{
|
|
"epoch": 7.69,
|
|
"learning_rate": 0.00027819548872180447,
|
|
"loss": 0.8823,
|
|
"step": 50
|
|
},
|
|
{
|
|
"epoch": 8.0,
|
|
"eval_accuracy": 0.5869565217391305,
|
|
"eval_loss": 1.0564936399459839,
|
|
"eval_runtime": 0.6594,
|
|
"eval_samples_per_second": 69.756,
|
|
"eval_steps_per_second": 3.033,
|
|
"step": 52
|
|
},
|
|
{
|
|
"epoch": 8.92,
|
|
"eval_accuracy": 0.7391304347826086,
|
|
"eval_loss": 0.941001296043396,
|
|
"eval_runtime": 0.6975,
|
|
"eval_samples_per_second": 65.947,
|
|
"eval_steps_per_second": 2.867,
|
|
"step": 58
|
|
},
|
|
{
|
|
"epoch": 9.23,
|
|
"learning_rate": 0.00027067669172932326,
|
|
"loss": 0.7637,
|
|
"step": 60
|
|
},
|
|
{
|
|
"epoch": 10.0,
|
|
"eval_accuracy": 0.7391304347826086,
|
|
"eval_loss": 0.9273515939712524,
|
|
"eval_runtime": 0.6226,
|
|
"eval_samples_per_second": 73.889,
|
|
"eval_steps_per_second": 3.213,
|
|
"step": 65
|
|
},
|
|
{
|
|
"epoch": 10.77,
|
|
"learning_rate": 0.00026315789473684205,
|
|
"loss": 0.6688,
|
|
"step": 70
|
|
},
|
|
{
|
|
"epoch": 10.92,
|
|
"eval_accuracy": 0.782608695652174,
|
|
"eval_loss": 0.8491992354393005,
|
|
"eval_runtime": 0.633,
|
|
"eval_samples_per_second": 72.669,
|
|
"eval_steps_per_second": 3.16,
|
|
"step": 71
|
|
},
|
|
{
|
|
"epoch": 12.0,
|
|
"eval_accuracy": 0.6739130434782609,
|
|
"eval_loss": 0.8905754089355469,
|
|
"eval_runtime": 0.651,
|
|
"eval_samples_per_second": 70.665,
|
|
"eval_steps_per_second": 3.072,
|
|
"step": 78
|
|
},
|
|
{
|
|
"epoch": 12.31,
|
|
"learning_rate": 0.0002556390977443609,
|
|
"loss": 0.5855,
|
|
"step": 80
|
|
},
|
|
{
|
|
"epoch": 12.92,
|
|
"eval_accuracy": 0.6521739130434783,
|
|
"eval_loss": 0.8928570747375488,
|
|
"eval_runtime": 0.7507,
|
|
"eval_samples_per_second": 61.274,
|
|
"eval_steps_per_second": 2.664,
|
|
"step": 84
|
|
},
|
|
{
|
|
"epoch": 13.85,
|
|
"learning_rate": 0.0002481203007518797,
|
|
"loss": 0.4921,
|
|
"step": 90
|
|
},
|
|
{
|
|
"epoch": 14.0,
|
|
"eval_accuracy": 0.7391304347826086,
|
|
"eval_loss": 0.8337807655334473,
|
|
"eval_runtime": 0.6453,
|
|
"eval_samples_per_second": 71.281,
|
|
"eval_steps_per_second": 3.099,
|
|
"step": 91
|
|
},
|
|
{
|
|
"epoch": 14.92,
|
|
"eval_accuracy": 0.782608695652174,
|
|
"eval_loss": 0.7686272263526917,
|
|
"eval_runtime": 0.6409,
|
|
"eval_samples_per_second": 71.778,
|
|
"eval_steps_per_second": 3.121,
|
|
"step": 97
|
|
},
|
|
{
|
|
"epoch": 15.38,
|
|
"learning_rate": 0.00024060150375939846,
|
|
"loss": 0.4318,
|
|
"step": 100
|
|
},
|
|
{
|
|
"epoch": 16.0,
|
|
"eval_accuracy": 0.7608695652173914,
|
|
"eval_loss": 0.8430270552635193,
|
|
"eval_runtime": 0.6391,
|
|
"eval_samples_per_second": 71.979,
|
|
"eval_steps_per_second": 3.13,
|
|
"step": 104
|
|
},
|
|
{
|
|
"epoch": 16.92,
|
|
"learning_rate": 0.00023308270676691725,
|
|
"loss": 0.386,
|
|
"step": 110
|
|
},
|
|
{
|
|
"epoch": 16.92,
|
|
"eval_accuracy": 0.782608695652174,
|
|
"eval_loss": 0.8314574956893921,
|
|
"eval_runtime": 0.6623,
|
|
"eval_samples_per_second": 69.453,
|
|
"eval_steps_per_second": 3.02,
|
|
"step": 110
|
|
},
|
|
{
|
|
"epoch": 18.0,
|
|
"eval_accuracy": 0.8260869565217391,
|
|
"eval_loss": 0.7668588757514954,
|
|
"eval_runtime": 0.6059,
|
|
"eval_samples_per_second": 75.926,
|
|
"eval_steps_per_second": 3.301,
|
|
"step": 117
|
|
},
|
|
{
|
|
"epoch": 18.46,
|
|
"learning_rate": 0.00022556390977443607,
|
|
"loss": 0.3483,
|
|
"step": 120
|
|
},
|
|
{
|
|
"epoch": 18.92,
|
|
"eval_accuracy": 0.717391304347826,
|
|
"eval_loss": 0.8347011208534241,
|
|
"eval_runtime": 0.6594,
|
|
"eval_samples_per_second": 69.764,
|
|
"eval_steps_per_second": 3.033,
|
|
"step": 123
|
|
},
|
|
{
|
|
"epoch": 20.0,
|
|
"learning_rate": 0.00021804511278195485,
|
|
"loss": 0.3023,
|
|
"step": 130
|
|
},
|
|
{
|
|
"epoch": 20.0,
|
|
"eval_accuracy": 0.6304347826086957,
|
|
"eval_loss": 1.1037073135375977,
|
|
"eval_runtime": 0.6056,
|
|
"eval_samples_per_second": 75.958,
|
|
"eval_steps_per_second": 3.303,
|
|
"step": 130
|
|
},
|
|
{
|
|
"epoch": 20.92,
|
|
"eval_accuracy": 0.717391304347826,
|
|
"eval_loss": 0.902391254901886,
|
|
"eval_runtime": 0.5892,
|
|
"eval_samples_per_second": 78.072,
|
|
"eval_steps_per_second": 3.394,
|
|
"step": 136
|
|
},
|
|
{
|
|
"epoch": 21.54,
|
|
"learning_rate": 0.00021052631578947364,
|
|
"loss": 0.2973,
|
|
"step": 140
|
|
},
|
|
{
|
|
"epoch": 22.0,
|
|
"eval_accuracy": 0.782608695652174,
|
|
"eval_loss": 0.7759842872619629,
|
|
"eval_runtime": 0.6372,
|
|
"eval_samples_per_second": 72.191,
|
|
"eval_steps_per_second": 3.139,
|
|
"step": 143
|
|
},
|
|
{
|
|
"epoch": 22.92,
|
|
"eval_accuracy": 0.782608695652174,
|
|
"eval_loss": 0.7400146722793579,
|
|
"eval_runtime": 0.6217,
|
|
"eval_samples_per_second": 73.985,
|
|
"eval_steps_per_second": 3.217,
|
|
"step": 149
|
|
},
|
|
{
|
|
"epoch": 23.08,
|
|
"learning_rate": 0.00020300751879699246,
|
|
"loss": 0.2529,
|
|
"step": 150
|
|
},
|
|
{
|
|
"epoch": 24.0,
|
|
"eval_accuracy": 0.717391304347826,
|
|
"eval_loss": 1.0057660341262817,
|
|
"eval_runtime": 0.6161,
|
|
"eval_samples_per_second": 74.665,
|
|
"eval_steps_per_second": 3.246,
|
|
"step": 156
|
|
},
|
|
{
|
|
"epoch": 24.62,
|
|
"learning_rate": 0.00019548872180451124,
|
|
"loss": 0.2086,
|
|
"step": 160
|
|
},
|
|
{
|
|
"epoch": 24.92,
|
|
"eval_accuracy": 0.7608695652173914,
|
|
"eval_loss": 0.9259914755821228,
|
|
"eval_runtime": 0.623,
|
|
"eval_samples_per_second": 73.836,
|
|
"eval_steps_per_second": 3.21,
|
|
"step": 162
|
|
},
|
|
{
|
|
"epoch": 26.0,
|
|
"eval_accuracy": 0.717391304347826,
|
|
"eval_loss": 0.8369772434234619,
|
|
"eval_runtime": 0.6104,
|
|
"eval_samples_per_second": 75.358,
|
|
"eval_steps_per_second": 3.276,
|
|
"step": 169
|
|
},
|
|
{
|
|
"epoch": 26.15,
|
|
"learning_rate": 0.00018796992481203006,
|
|
"loss": 0.2265,
|
|
"step": 170
|
|
},
|
|
{
|
|
"epoch": 26.92,
|
|
"eval_accuracy": 0.7391304347826086,
|
|
"eval_loss": 0.8060359954833984,
|
|
"eval_runtime": 0.6039,
|
|
"eval_samples_per_second": 76.166,
|
|
"eval_steps_per_second": 3.312,
|
|
"step": 175
|
|
},
|
|
{
|
|
"epoch": 27.69,
|
|
"learning_rate": 0.00018045112781954885,
|
|
"loss": 0.1942,
|
|
"step": 180
|
|
},
|
|
{
|
|
"epoch": 28.0,
|
|
"eval_accuracy": 0.6956521739130435,
|
|
"eval_loss": 0.981198787689209,
|
|
"eval_runtime": 0.6812,
|
|
"eval_samples_per_second": 67.531,
|
|
"eval_steps_per_second": 2.936,
|
|
"step": 182
|
|
},
|
|
{
|
|
"epoch": 28.92,
|
|
"eval_accuracy": 0.7391304347826086,
|
|
"eval_loss": 0.89959317445755,
|
|
"eval_runtime": 0.629,
|
|
"eval_samples_per_second": 73.137,
|
|
"eval_steps_per_second": 3.18,
|
|
"step": 188
|
|
},
|
|
{
|
|
"epoch": 29.23,
|
|
"learning_rate": 0.00017293233082706764,
|
|
"loss": 0.1708,
|
|
"step": 190
|
|
},
|
|
{
|
|
"epoch": 30.0,
|
|
"eval_accuracy": 0.6956521739130435,
|
|
"eval_loss": 0.9629535675048828,
|
|
"eval_runtime": 0.6266,
|
|
"eval_samples_per_second": 73.406,
|
|
"eval_steps_per_second": 3.192,
|
|
"step": 195
|
|
},
|
|
{
|
|
"epoch": 30.77,
|
|
"learning_rate": 0.00016541353383458645,
|
|
"loss": 0.1747,
|
|
"step": 200
|
|
},
|
|
{
|
|
"epoch": 30.92,
|
|
"eval_accuracy": 0.717391304347826,
|
|
"eval_loss": 0.9691163897514343,
|
|
"eval_runtime": 0.7078,
|
|
"eval_samples_per_second": 64.986,
|
|
"eval_steps_per_second": 2.825,
|
|
"step": 201
|
|
},
|
|
{
|
|
"epoch": 32.0,
|
|
"eval_accuracy": 0.7391304347826086,
|
|
"eval_loss": 1.0016721487045288,
|
|
"eval_runtime": 0.6466,
|
|
"eval_samples_per_second": 71.146,
|
|
"eval_steps_per_second": 3.093,
|
|
"step": 208
|
|
},
|
|
{
|
|
"epoch": 32.31,
|
|
"learning_rate": 0.00015789473684210524,
|
|
"loss": 0.1461,
|
|
"step": 210
|
|
},
|
|
{
|
|
"epoch": 32.92,
|
|
"eval_accuracy": 0.6956521739130435,
|
|
"eval_loss": 0.9964732527732849,
|
|
"eval_runtime": 0.6084,
|
|
"eval_samples_per_second": 75.606,
|
|
"eval_steps_per_second": 3.287,
|
|
"step": 214
|
|
},
|
|
{
|
|
"epoch": 33.85,
|
|
"learning_rate": 0.00015037593984962403,
|
|
"loss": 0.1457,
|
|
"step": 220
|
|
},
|
|
{
|
|
"epoch": 34.0,
|
|
"eval_accuracy": 0.7391304347826086,
|
|
"eval_loss": 0.950577437877655,
|
|
"eval_runtime": 0.5862,
|
|
"eval_samples_per_second": 78.471,
|
|
"eval_steps_per_second": 3.412,
|
|
"step": 221
|
|
},
|
|
{
|
|
"epoch": 34.92,
|
|
"eval_accuracy": 0.7391304347826086,
|
|
"eval_loss": 0.9834409952163696,
|
|
"eval_runtime": 0.5974,
|
|
"eval_samples_per_second": 77.005,
|
|
"eval_steps_per_second": 3.348,
|
|
"step": 227
|
|
},
|
|
{
|
|
"epoch": 35.38,
|
|
"learning_rate": 0.00014285714285714284,
|
|
"loss": 0.1814,
|
|
"step": 230
|
|
},
|
|
{
|
|
"epoch": 36.0,
|
|
"eval_accuracy": 0.7608695652173914,
|
|
"eval_loss": 1.0190751552581787,
|
|
"eval_runtime": 0.5969,
|
|
"eval_samples_per_second": 77.071,
|
|
"eval_steps_per_second": 3.351,
|
|
"step": 234
|
|
},
|
|
{
|
|
"epoch": 36.92,
|
|
"learning_rate": 0.00013533834586466163,
|
|
"loss": 0.1383,
|
|
"step": 240
|
|
},
|
|
{
|
|
"epoch": 36.92,
|
|
"eval_accuracy": 0.7608695652173914,
|
|
"eval_loss": 0.8806793093681335,
|
|
"eval_runtime": 0.6184,
|
|
"eval_samples_per_second": 74.383,
|
|
"eval_steps_per_second": 3.234,
|
|
"step": 240
|
|
},
|
|
{
|
|
"epoch": 38.0,
|
|
"eval_accuracy": 0.7608695652173914,
|
|
"eval_loss": 0.8724012970924377,
|
|
"eval_runtime": 0.7088,
|
|
"eval_samples_per_second": 64.9,
|
|
"eval_steps_per_second": 2.822,
|
|
"step": 247
|
|
},
|
|
{
|
|
"epoch": 38.46,
|
|
"learning_rate": 0.00012781954887218044,
|
|
"loss": 0.1718,
|
|
"step": 250
|
|
},
|
|
{
|
|
"epoch": 38.92,
|
|
"eval_accuracy": 0.7391304347826086,
|
|
"eval_loss": 0.8089635968208313,
|
|
"eval_runtime": 0.6027,
|
|
"eval_samples_per_second": 76.318,
|
|
"eval_steps_per_second": 3.318,
|
|
"step": 253
|
|
},
|
|
{
|
|
"epoch": 40.0,
|
|
"learning_rate": 0.00012030075187969923,
|
|
"loss": 0.1289,
|
|
"step": 260
|
|
},
|
|
{
|
|
"epoch": 40.0,
|
|
"eval_accuracy": 0.7608695652173914,
|
|
"eval_loss": 0.8709278702735901,
|
|
"eval_runtime": 0.5996,
|
|
"eval_samples_per_second": 76.718,
|
|
"eval_steps_per_second": 3.336,
|
|
"step": 260
|
|
},
|
|
{
|
|
"epoch": 40.92,
|
|
"eval_accuracy": 0.7391304347826086,
|
|
"eval_loss": 0.9704360961914062,
|
|
"eval_runtime": 0.6075,
|
|
"eval_samples_per_second": 75.717,
|
|
"eval_steps_per_second": 3.292,
|
|
"step": 266
|
|
},
|
|
{
|
|
"epoch": 41.54,
|
|
"learning_rate": 0.00011278195488721803,
|
|
"loss": 0.1193,
|
|
"step": 270
|
|
},
|
|
{
|
|
"epoch": 42.0,
|
|
"eval_accuracy": 0.7391304347826086,
|
|
"eval_loss": 1.0517703294754028,
|
|
"eval_runtime": 0.6432,
|
|
"eval_samples_per_second": 71.516,
|
|
"eval_steps_per_second": 3.109,
|
|
"step": 273
|
|
},
|
|
{
|
|
"epoch": 42.92,
|
|
"eval_accuracy": 0.717391304347826,
|
|
"eval_loss": 1.0156539678573608,
|
|
"eval_runtime": 0.5992,
|
|
"eval_samples_per_second": 76.773,
|
|
"eval_steps_per_second": 3.338,
|
|
"step": 279
|
|
},
|
|
{
|
|
"epoch": 43.08,
|
|
"learning_rate": 0.00010526315789473682,
|
|
"loss": 0.1224,
|
|
"step": 280
|
|
},
|
|
{
|
|
"epoch": 44.0,
|
|
"eval_accuracy": 0.7391304347826086,
|
|
"eval_loss": 1.0793591737747192,
|
|
"eval_runtime": 0.6282,
|
|
"eval_samples_per_second": 73.225,
|
|
"eval_steps_per_second": 3.184,
|
|
"step": 286
|
|
},
|
|
{
|
|
"epoch": 44.62,
|
|
"learning_rate": 9.774436090225562e-05,
|
|
"loss": 0.1104,
|
|
"step": 290
|
|
},
|
|
{
|
|
"epoch": 44.92,
|
|
"eval_accuracy": 0.7391304347826086,
|
|
"eval_loss": 1.040244698524475,
|
|
"eval_runtime": 0.6284,
|
|
"eval_samples_per_second": 73.198,
|
|
"eval_steps_per_second": 3.183,
|
|
"step": 292
|
|
},
|
|
{
|
|
"epoch": 46.0,
|
|
"eval_accuracy": 0.7608695652173914,
|
|
"eval_loss": 0.9836814999580383,
|
|
"eval_runtime": 0.5971,
|
|
"eval_samples_per_second": 77.042,
|
|
"eval_steps_per_second": 3.35,
|
|
"step": 299
|
|
},
|
|
{
|
|
"epoch": 46.15,
|
|
"learning_rate": 9.022556390977442e-05,
|
|
"loss": 0.1055,
|
|
"step": 300
|
|
},
|
|
{
|
|
"epoch": 46.92,
|
|
"eval_accuracy": 0.717391304347826,
|
|
"eval_loss": 1.0322998762130737,
|
|
"eval_runtime": 0.6274,
|
|
"eval_samples_per_second": 73.316,
|
|
"eval_steps_per_second": 3.188,
|
|
"step": 305
|
|
},
|
|
{
|
|
"epoch": 47.69,
|
|
"learning_rate": 8.270676691729323e-05,
|
|
"loss": 0.1242,
|
|
"step": 310
|
|
},
|
|
{
|
|
"epoch": 48.0,
|
|
"eval_accuracy": 0.7391304347826086,
|
|
"eval_loss": 0.9906932711601257,
|
|
"eval_runtime": 0.6469,
|
|
"eval_samples_per_second": 71.108,
|
|
"eval_steps_per_second": 3.092,
|
|
"step": 312
|
|
},
|
|
{
|
|
"epoch": 48.92,
|
|
"eval_accuracy": 0.7608695652173914,
|
|
"eval_loss": 1.043601155281067,
|
|
"eval_runtime": 0.6539,
|
|
"eval_samples_per_second": 70.351,
|
|
"eval_steps_per_second": 3.059,
|
|
"step": 318
|
|
},
|
|
{
|
|
"epoch": 49.23,
|
|
"learning_rate": 7.518796992481201e-05,
|
|
"loss": 0.1283,
|
|
"step": 320
|
|
},
|
|
{
|
|
"epoch": 50.0,
|
|
"eval_accuracy": 0.7391304347826086,
|
|
"eval_loss": 0.9828565716743469,
|
|
"eval_runtime": 0.6221,
|
|
"eval_samples_per_second": 73.942,
|
|
"eval_steps_per_second": 3.215,
|
|
"step": 325
|
|
},
|
|
{
|
|
"epoch": 50.77,
|
|
"learning_rate": 6.766917293233081e-05,
|
|
"loss": 0.1035,
|
|
"step": 330
|
|
},
|
|
{
|
|
"epoch": 50.92,
|
|
"eval_accuracy": 0.717391304347826,
|
|
"eval_loss": 1.040035367012024,
|
|
"eval_runtime": 0.6055,
|
|
"eval_samples_per_second": 75.973,
|
|
"eval_steps_per_second": 3.303,
|
|
"step": 331
|
|
},
|
|
{
|
|
"epoch": 52.0,
|
|
"eval_accuracy": 0.717391304347826,
|
|
"eval_loss": 1.0413511991500854,
|
|
"eval_runtime": 0.6051,
|
|
"eval_samples_per_second": 76.025,
|
|
"eval_steps_per_second": 3.305,
|
|
"step": 338
|
|
},
|
|
{
|
|
"epoch": 52.31,
|
|
"learning_rate": 6.0150375939849616e-05,
|
|
"loss": 0.1066,
|
|
"step": 340
|
|
},
|
|
{
|
|
"epoch": 52.92,
|
|
"eval_accuracy": 0.6956521739130435,
|
|
"eval_loss": 1.0958248376846313,
|
|
"eval_runtime": 0.6107,
|
|
"eval_samples_per_second": 75.32,
|
|
"eval_steps_per_second": 3.275,
|
|
"step": 344
|
|
},
|
|
{
|
|
"epoch": 53.85,
|
|
"learning_rate": 5.263157894736841e-05,
|
|
"loss": 0.0863,
|
|
"step": 350
|
|
},
|
|
{
|
|
"epoch": 54.0,
|
|
"eval_accuracy": 0.717391304347826,
|
|
"eval_loss": 1.0913535356521606,
|
|
"eval_runtime": 0.6053,
|
|
"eval_samples_per_second": 76.0,
|
|
"eval_steps_per_second": 3.304,
|
|
"step": 351
|
|
},
|
|
{
|
|
"epoch": 54.92,
|
|
"eval_accuracy": 0.717391304347826,
|
|
"eval_loss": 1.081627607345581,
|
|
"eval_runtime": 0.6407,
|
|
"eval_samples_per_second": 71.791,
|
|
"eval_steps_per_second": 3.121,
|
|
"step": 357
|
|
},
|
|
{
|
|
"epoch": 55.38,
|
|
"learning_rate": 4.511278195488721e-05,
|
|
"loss": 0.1062,
|
|
"step": 360
|
|
},
|
|
{
|
|
"epoch": 56.0,
|
|
"eval_accuracy": 0.717391304347826,
|
|
"eval_loss": 1.0086578130722046,
|
|
"eval_runtime": 0.642,
|
|
"eval_samples_per_second": 71.655,
|
|
"eval_steps_per_second": 3.115,
|
|
"step": 364
|
|
},
|
|
{
|
|
"epoch": 56.92,
|
|
"learning_rate": 3.7593984962406006e-05,
|
|
"loss": 0.1214,
|
|
"step": 370
|
|
},
|
|
{
|
|
"epoch": 56.92,
|
|
"eval_accuracy": 0.7391304347826086,
|
|
"eval_loss": 1.0453943014144897,
|
|
"eval_runtime": 0.6492,
|
|
"eval_samples_per_second": 70.854,
|
|
"eval_steps_per_second": 3.081,
|
|
"step": 370
|
|
},
|
|
{
|
|
"epoch": 58.0,
|
|
"eval_accuracy": 0.7391304347826086,
|
|
"eval_loss": 1.0324147939682007,
|
|
"eval_runtime": 0.599,
|
|
"eval_samples_per_second": 76.796,
|
|
"eval_steps_per_second": 3.339,
|
|
"step": 377
|
|
},
|
|
{
|
|
"epoch": 58.46,
|
|
"learning_rate": 3.0075187969924808e-05,
|
|
"loss": 0.0984,
|
|
"step": 380
|
|
},
|
|
{
|
|
"epoch": 58.92,
|
|
"eval_accuracy": 0.6739130434782609,
|
|
"eval_loss": 1.0590661764144897,
|
|
"eval_runtime": 0.6829,
|
|
"eval_samples_per_second": 67.355,
|
|
"eval_steps_per_second": 2.928,
|
|
"step": 383
|
|
},
|
|
{
|
|
"epoch": 60.0,
|
|
"learning_rate": 2.2556390977443606e-05,
|
|
"loss": 0.0966,
|
|
"step": 390
|
|
},
|
|
{
|
|
"epoch": 60.0,
|
|
"eval_accuracy": 0.6956521739130435,
|
|
"eval_loss": 1.0037105083465576,
|
|
"eval_runtime": 0.6261,
|
|
"eval_samples_per_second": 73.471,
|
|
"eval_steps_per_second": 3.194,
|
|
"step": 390
|
|
},
|
|
{
|
|
"epoch": 60.92,
|
|
"eval_accuracy": 0.6956521739130435,
|
|
"eval_loss": 0.9887439608573914,
|
|
"eval_runtime": 0.64,
|
|
"eval_samples_per_second": 71.875,
|
|
"eval_steps_per_second": 3.125,
|
|
"step": 396
|
|
},
|
|
{
|
|
"epoch": 61.54,
|
|
"learning_rate": 1.5037593984962404e-05,
|
|
"loss": 0.0626,
|
|
"step": 400
|
|
},
|
|
{
|
|
"epoch": 62.0,
|
|
"eval_accuracy": 0.6739130434782609,
|
|
"eval_loss": 1.0294169187545776,
|
|
"eval_runtime": 0.5869,
|
|
"eval_samples_per_second": 78.378,
|
|
"eval_steps_per_second": 3.408,
|
|
"step": 403
|
|
},
|
|
{
|
|
"epoch": 62.92,
|
|
"eval_accuracy": 0.717391304347826,
|
|
"eval_loss": 0.9938833713531494,
|
|
"eval_runtime": 0.6096,
|
|
"eval_samples_per_second": 75.462,
|
|
"eval_steps_per_second": 3.281,
|
|
"step": 409
|
|
},
|
|
{
|
|
"epoch": 63.08,
|
|
"learning_rate": 7.518796992481202e-06,
|
|
"loss": 0.085,
|
|
"step": 410
|
|
},
|
|
{
|
|
"epoch": 64.0,
|
|
"eval_accuracy": 0.6956521739130435,
|
|
"eval_loss": 0.9885643124580383,
|
|
"eval_runtime": 0.6332,
|
|
"eval_samples_per_second": 72.646,
|
|
"eval_steps_per_second": 3.159,
|
|
"step": 416
|
|
},
|
|
{
|
|
"epoch": 64.62,
|
|
"learning_rate": 0.0,
|
|
"loss": 0.068,
|
|
"step": 420
|
|
},
|
|
{
|
|
"epoch": 64.62,
|
|
"eval_accuracy": 0.6739130434782609,
|
|
"eval_loss": 1.0773022174835205,
|
|
"eval_runtime": 0.6376,
|
|
"eval_samples_per_second": 72.141,
|
|
"eval_steps_per_second": 3.137,
|
|
"step": 420
|
|
},
|
|
{
|
|
"epoch": 64.62,
|
|
"step": 420,
|
|
"total_flos": 1.450878471367557e+17,
|
|
"train_loss": 0.3280091051544462,
|
|
"train_runtime": 248.4253,
|
|
"train_samples_per_second": 230.774,
|
|
"train_steps_per_second": 1.691
|
|
}
|
|
],
|
|
"logging_steps": 10,
|
|
"max_steps": 420,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 70,
|
|
"save_steps": 500,
|
|
"total_flos": 1.450878471367557e+17,
|
|
"train_batch_size": 32,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|
|
|