sentinelv2 / checkpoint-162 /trainer_state.json
Alignment-Lab-AI's picture
Training in progress, epoch 2, checkpoint
9a65339
raw
history blame
21.2 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.0253164556962027,
"eval_steps": 20,
"global_step": 162,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 0.0,
"loss": 0.6625,
"step": 1
},
{
"epoch": 0.01,
"eval_loss": 0.9669284820556641,
"eval_runtime": 22.2599,
"eval_samples_per_second": 11.231,
"eval_steps_per_second": 2.83,
"step": 1
},
{
"epoch": 0.03,
"learning_rate": 6.666666e-07,
"loss": 0.7012,
"step": 2
},
{
"epoch": 0.04,
"learning_rate": 1.3333332e-06,
"loss": 0.6954,
"step": 3
},
{
"epoch": 0.05,
"learning_rate": 1.9999997999999996e-06,
"loss": 0.6383,
"step": 4
},
{
"epoch": 0.06,
"learning_rate": 2.6666664e-06,
"loss": 0.6993,
"step": 5
},
{
"epoch": 0.08,
"learning_rate": 3.3333329999999998e-06,
"loss": 0.6388,
"step": 6
},
{
"epoch": 0.09,
"learning_rate": 3.999999599999999e-06,
"loss": 0.6847,
"step": 7
},
{
"epoch": 0.1,
"learning_rate": 4.6666662e-06,
"loss": 0.6973,
"step": 8
},
{
"epoch": 0.11,
"learning_rate": 5.3333328e-06,
"loss": 0.703,
"step": 9
},
{
"epoch": 0.13,
"learning_rate": 5.999999399999999e-06,
"loss": 0.6826,
"step": 10
},
{
"epoch": 0.14,
"learning_rate": 6.6666659999999995e-06,
"loss": 0.6939,
"step": 11
},
{
"epoch": 0.15,
"learning_rate": 7.333332599999999e-06,
"loss": 0.6879,
"step": 12
},
{
"epoch": 0.16,
"learning_rate": 7.999999199999998e-06,
"loss": 0.687,
"step": 13
},
{
"epoch": 0.18,
"learning_rate": 8.6666658e-06,
"loss": 0.6471,
"step": 14
},
{
"epoch": 0.19,
"learning_rate": 9.3333324e-06,
"loss": 0.6658,
"step": 15
},
{
"epoch": 0.2,
"learning_rate": 9.999998999999998e-06,
"loss": 0.6669,
"step": 16
},
{
"epoch": 0.22,
"learning_rate": 1.06666656e-05,
"loss": 0.6451,
"step": 17
},
{
"epoch": 0.23,
"learning_rate": 1.13333322e-05,
"loss": 0.6558,
"step": 18
},
{
"epoch": 0.24,
"learning_rate": 1.1999998799999998e-05,
"loss": 0.6714,
"step": 19
},
{
"epoch": 0.25,
"learning_rate": 1.26666654e-05,
"loss": 0.7123,
"step": 20
},
{
"epoch": 0.25,
"eval_loss": 0.9579811096191406,
"eval_runtime": 22.7924,
"eval_samples_per_second": 10.969,
"eval_steps_per_second": 2.764,
"step": 20
},
{
"epoch": 0.27,
"learning_rate": 1.3333331999999999e-05,
"loss": 0.6388,
"step": 21
},
{
"epoch": 0.28,
"learning_rate": 1.3999998599999999e-05,
"loss": 0.6355,
"step": 22
},
{
"epoch": 0.29,
"learning_rate": 1.4666665199999999e-05,
"loss": 0.6237,
"step": 23
},
{
"epoch": 0.3,
"learning_rate": 1.53333318e-05,
"loss": 0.6214,
"step": 24
},
{
"epoch": 0.32,
"learning_rate": 1.5999998399999997e-05,
"loss": 0.6045,
"step": 25
},
{
"epoch": 0.33,
"learning_rate": 1.6666665e-05,
"loss": 0.671,
"step": 26
},
{
"epoch": 0.34,
"learning_rate": 1.73333316e-05,
"loss": 0.6042,
"step": 27
},
{
"epoch": 0.35,
"learning_rate": 1.7999998199999998e-05,
"loss": 0.6722,
"step": 28
},
{
"epoch": 0.37,
"learning_rate": 1.86666648e-05,
"loss": 0.6132,
"step": 29
},
{
"epoch": 0.38,
"learning_rate": 1.9333331399999998e-05,
"loss": 0.5651,
"step": 30
},
{
"epoch": 0.39,
"learning_rate": 1.9999997999999996e-05,
"loss": 0.6525,
"step": 31
},
{
"epoch": 0.41,
"learning_rate": 2.0666664599999998e-05,
"loss": 0.6532,
"step": 32
},
{
"epoch": 0.42,
"learning_rate": 2.13333312e-05,
"loss": 0.6167,
"step": 33
},
{
"epoch": 0.43,
"learning_rate": 2.1999997799999997e-05,
"loss": 0.607,
"step": 34
},
{
"epoch": 0.44,
"learning_rate": 2.26666644e-05,
"loss": 0.5843,
"step": 35
},
{
"epoch": 0.46,
"learning_rate": 2.3333330999999997e-05,
"loss": 0.6093,
"step": 36
},
{
"epoch": 0.47,
"learning_rate": 2.3999997599999995e-05,
"loss": 0.5967,
"step": 37
},
{
"epoch": 0.48,
"learning_rate": 2.4666664199999997e-05,
"loss": 0.6015,
"step": 38
},
{
"epoch": 0.49,
"learning_rate": 2.53333308e-05,
"loss": 0.6263,
"step": 39
},
{
"epoch": 0.51,
"learning_rate": 2.59999974e-05,
"loss": 0.5923,
"step": 40
},
{
"epoch": 0.51,
"eval_loss": 0.9467829465866089,
"eval_runtime": 22.6276,
"eval_samples_per_second": 11.048,
"eval_steps_per_second": 2.784,
"step": 40
},
{
"epoch": 0.52,
"learning_rate": 2.6666663999999998e-05,
"loss": 0.5801,
"step": 41
},
{
"epoch": 0.53,
"learning_rate": 2.7333330599999996e-05,
"loss": 0.576,
"step": 42
},
{
"epoch": 0.54,
"learning_rate": 2.7999997199999998e-05,
"loss": 0.6317,
"step": 43
},
{
"epoch": 0.56,
"learning_rate": 2.8666663799999996e-05,
"loss": 0.5874,
"step": 44
},
{
"epoch": 0.57,
"learning_rate": 2.9333330399999998e-05,
"loss": 0.6109,
"step": 45
},
{
"epoch": 0.58,
"learning_rate": 2.9999997e-05,
"loss": 0.5274,
"step": 46
},
{
"epoch": 0.59,
"learning_rate": 3.06666636e-05,
"loss": 0.5698,
"step": 47
},
{
"epoch": 0.61,
"learning_rate": 3.133333019999999e-05,
"loss": 0.544,
"step": 48
},
{
"epoch": 0.62,
"learning_rate": 3.1999996799999994e-05,
"loss": 0.5586,
"step": 49
},
{
"epoch": 0.63,
"learning_rate": 3.2666663399999995e-05,
"loss": 0.5666,
"step": 50
},
{
"epoch": 0.65,
"learning_rate": 3.333333e-05,
"loss": 0.586,
"step": 51
},
{
"epoch": 0.66,
"learning_rate": 3.39999966e-05,
"loss": 0.6225,
"step": 52
},
{
"epoch": 0.67,
"learning_rate": 3.46666632e-05,
"loss": 0.5588,
"step": 53
},
{
"epoch": 0.68,
"learning_rate": 3.53333298e-05,
"loss": 0.5705,
"step": 54
},
{
"epoch": 0.7,
"learning_rate": 3.5999996399999996e-05,
"loss": 0.5628,
"step": 55
},
{
"epoch": 0.71,
"learning_rate": 3.6666663e-05,
"loss": 0.5868,
"step": 56
},
{
"epoch": 0.72,
"learning_rate": 3.73333296e-05,
"loss": 0.5583,
"step": 57
},
{
"epoch": 0.73,
"learning_rate": 3.7999996199999994e-05,
"loss": 0.5858,
"step": 58
},
{
"epoch": 0.75,
"learning_rate": 3.8666662799999996e-05,
"loss": 0.5756,
"step": 59
},
{
"epoch": 0.76,
"learning_rate": 3.93333294e-05,
"loss": 0.5822,
"step": 60
},
{
"epoch": 0.76,
"eval_loss": 0.9389348030090332,
"eval_runtime": 22.5892,
"eval_samples_per_second": 11.067,
"eval_steps_per_second": 2.789,
"step": 60
},
{
"epoch": 0.77,
"learning_rate": 3.999999599999999e-05,
"loss": 0.5667,
"step": 61
},
{
"epoch": 0.78,
"learning_rate": 4.0666662599999994e-05,
"loss": 0.5616,
"step": 62
},
{
"epoch": 0.8,
"learning_rate": 4.1333329199999995e-05,
"loss": 0.5678,
"step": 63
},
{
"epoch": 0.81,
"learning_rate": 4.19999958e-05,
"loss": 0.5514,
"step": 64
},
{
"epoch": 0.82,
"learning_rate": 4.26666624e-05,
"loss": 0.5732,
"step": 65
},
{
"epoch": 0.84,
"learning_rate": 4.3333329e-05,
"loss": 0.5794,
"step": 66
},
{
"epoch": 0.85,
"learning_rate": 4.3999995599999995e-05,
"loss": 0.5582,
"step": 67
},
{
"epoch": 0.86,
"learning_rate": 4.4666662199999996e-05,
"loss": 0.5575,
"step": 68
},
{
"epoch": 0.87,
"learning_rate": 4.53333288e-05,
"loss": 0.5558,
"step": 69
},
{
"epoch": 0.89,
"learning_rate": 4.599999539999999e-05,
"loss": 0.5707,
"step": 70
},
{
"epoch": 0.9,
"learning_rate": 4.6666661999999994e-05,
"loss": 0.54,
"step": 71
},
{
"epoch": 0.91,
"learning_rate": 4.7333328599999996e-05,
"loss": 0.5435,
"step": 72
},
{
"epoch": 0.92,
"learning_rate": 4.799999519999999e-05,
"loss": 0.5649,
"step": 73
},
{
"epoch": 0.94,
"learning_rate": 4.866666179999999e-05,
"loss": 0.5617,
"step": 74
},
{
"epoch": 0.95,
"learning_rate": 4.9333328399999994e-05,
"loss": 0.5651,
"step": 75
},
{
"epoch": 0.96,
"learning_rate": 4.9999994999999995e-05,
"loss": 0.5113,
"step": 76
},
{
"epoch": 0.97,
"learning_rate": 5.06666616e-05,
"loss": 0.5389,
"step": 77
},
{
"epoch": 0.99,
"learning_rate": 5.13333282e-05,
"loss": 0.5162,
"step": 78
},
{
"epoch": 1.0,
"learning_rate": 5.19999948e-05,
"loss": 0.5397,
"step": 79
},
{
"epoch": 1.01,
"learning_rate": 5.2666661399999995e-05,
"loss": 0.5794,
"step": 80
},
{
"epoch": 1.01,
"eval_loss": 0.9314417839050293,
"eval_runtime": 22.7753,
"eval_samples_per_second": 10.977,
"eval_steps_per_second": 2.766,
"step": 80
},
{
"epoch": 1.03,
"learning_rate": 5.3333327999999996e-05,
"loss": 0.5263,
"step": 81
},
{
"epoch": 1.01,
"learning_rate": 5.39999946e-05,
"loss": 0.5141,
"step": 82
},
{
"epoch": 1.03,
"learning_rate": 5.466666119999999e-05,
"loss": 0.5573,
"step": 83
},
{
"epoch": 1.04,
"learning_rate": 5.5333327799999994e-05,
"loss": 0.5629,
"step": 84
},
{
"epoch": 1.05,
"learning_rate": 5.5999994399999996e-05,
"loss": 0.5043,
"step": 85
},
{
"epoch": 1.06,
"learning_rate": 5.666666099999999e-05,
"loss": 0.562,
"step": 86
},
{
"epoch": 1.08,
"learning_rate": 5.733332759999999e-05,
"loss": 0.4978,
"step": 87
},
{
"epoch": 1.09,
"learning_rate": 5.7999994199999994e-05,
"loss": 0.5502,
"step": 88
},
{
"epoch": 1.1,
"learning_rate": 5.8666660799999995e-05,
"loss": 0.5499,
"step": 89
},
{
"epoch": 1.11,
"learning_rate": 5.93333274e-05,
"loss": 0.5547,
"step": 90
},
{
"epoch": 1.13,
"learning_rate": 5.9999994e-05,
"loss": 0.5452,
"step": 91
},
{
"epoch": 1.14,
"learning_rate": 6.066666059999999e-05,
"loss": 0.5544,
"step": 92
},
{
"epoch": 1.15,
"learning_rate": 6.13333272e-05,
"loss": 0.5483,
"step": 93
},
{
"epoch": 1.16,
"learning_rate": 6.19999938e-05,
"loss": 0.5641,
"step": 94
},
{
"epoch": 1.18,
"learning_rate": 6.266666039999998e-05,
"loss": 0.5316,
"step": 95
},
{
"epoch": 1.19,
"learning_rate": 6.333332699999999e-05,
"loss": 0.526,
"step": 96
},
{
"epoch": 1.2,
"learning_rate": 6.399999359999999e-05,
"loss": 0.5443,
"step": 97
},
{
"epoch": 1.22,
"learning_rate": 6.46666602e-05,
"loss": 0.5111,
"step": 98
},
{
"epoch": 1.23,
"learning_rate": 6.533332679999999e-05,
"loss": 0.5298,
"step": 99
},
{
"epoch": 1.24,
"learning_rate": 6.59999934e-05,
"loss": 0.5431,
"step": 100
},
{
"epoch": 1.24,
"eval_loss": 0.9199196100234985,
"eval_runtime": 22.8496,
"eval_samples_per_second": 10.941,
"eval_steps_per_second": 2.757,
"step": 100
},
{
"epoch": 1.25,
"learning_rate": 6.666666e-05,
"loss": 0.585,
"step": 101
},
{
"epoch": 1.27,
"learning_rate": 6.64406713220339e-05,
"loss": 0.5143,
"step": 102
},
{
"epoch": 1.28,
"learning_rate": 6.621468264406779e-05,
"loss": 0.5061,
"step": 103
},
{
"epoch": 1.29,
"learning_rate": 6.59886939661017e-05,
"loss": 0.5098,
"step": 104
},
{
"epoch": 1.3,
"learning_rate": 6.576270528813559e-05,
"loss": 0.5188,
"step": 105
},
{
"epoch": 1.32,
"learning_rate": 6.553671661016948e-05,
"loss": 0.5005,
"step": 106
},
{
"epoch": 1.33,
"learning_rate": 6.531072793220339e-05,
"loss": 0.5709,
"step": 107
},
{
"epoch": 1.34,
"learning_rate": 6.508473925423728e-05,
"loss": 0.4985,
"step": 108
},
{
"epoch": 1.35,
"learning_rate": 6.485875057627117e-05,
"loss": 0.5767,
"step": 109
},
{
"epoch": 1.37,
"learning_rate": 6.463276189830508e-05,
"loss": 0.517,
"step": 110
},
{
"epoch": 1.38,
"learning_rate": 6.440677322033897e-05,
"loss": 0.4759,
"step": 111
},
{
"epoch": 1.39,
"learning_rate": 6.418078454237288e-05,
"loss": 0.5536,
"step": 112
},
{
"epoch": 1.41,
"learning_rate": 6.395479586440677e-05,
"loss": 0.5657,
"step": 113
},
{
"epoch": 1.42,
"learning_rate": 6.372880718644068e-05,
"loss": 0.525,
"step": 114
},
{
"epoch": 1.43,
"learning_rate": 6.350281850847457e-05,
"loss": 0.5113,
"step": 115
},
{
"epoch": 1.44,
"learning_rate": 6.327682983050848e-05,
"loss": 0.4869,
"step": 116
},
{
"epoch": 1.46,
"learning_rate": 6.305084115254237e-05,
"loss": 0.5074,
"step": 117
},
{
"epoch": 1.47,
"learning_rate": 6.282485247457626e-05,
"loss": 0.5043,
"step": 118
},
{
"epoch": 1.48,
"learning_rate": 6.259886379661017e-05,
"loss": 0.5116,
"step": 119
},
{
"epoch": 1.49,
"learning_rate": 6.237287511864406e-05,
"loss": 0.5317,
"step": 120
},
{
"epoch": 1.49,
"eval_loss": 0.9122854471206665,
"eval_runtime": 22.8969,
"eval_samples_per_second": 10.919,
"eval_steps_per_second": 2.751,
"step": 120
},
{
"epoch": 1.51,
"learning_rate": 6.214688644067795e-05,
"loss": 0.5049,
"step": 121
},
{
"epoch": 1.52,
"learning_rate": 6.192089776271186e-05,
"loss": 0.496,
"step": 122
},
{
"epoch": 1.53,
"learning_rate": 6.169490908474575e-05,
"loss": 0.4951,
"step": 123
},
{
"epoch": 1.54,
"learning_rate": 6.146892040677966e-05,
"loss": 0.5489,
"step": 124
},
{
"epoch": 1.56,
"learning_rate": 6.124293172881355e-05,
"loss": 0.5071,
"step": 125
},
{
"epoch": 1.57,
"learning_rate": 6.1016943050847455e-05,
"loss": 0.5312,
"step": 126
},
{
"epoch": 1.58,
"learning_rate": 6.079095437288135e-05,
"loss": 0.4442,
"step": 127
},
{
"epoch": 1.59,
"learning_rate": 6.056496569491525e-05,
"loss": 0.4944,
"step": 128
},
{
"epoch": 1.61,
"learning_rate": 6.033897701694915e-05,
"loss": 0.4615,
"step": 129
},
{
"epoch": 1.62,
"learning_rate": 6.0112988338983045e-05,
"loss": 0.4855,
"step": 130
},
{
"epoch": 1.63,
"learning_rate": 5.9886999661016945e-05,
"loss": 0.4862,
"step": 131
},
{
"epoch": 1.65,
"learning_rate": 5.9661010983050844e-05,
"loss": 0.521,
"step": 132
},
{
"epoch": 1.66,
"learning_rate": 5.943502230508474e-05,
"loss": 0.5537,
"step": 133
},
{
"epoch": 1.67,
"learning_rate": 5.9209033627118636e-05,
"loss": 0.4828,
"step": 134
},
{
"epoch": 1.68,
"learning_rate": 5.898304494915254e-05,
"loss": 0.4932,
"step": 135
},
{
"epoch": 1.7,
"learning_rate": 5.8757056271186434e-05,
"loss": 0.4909,
"step": 136
},
{
"epoch": 1.71,
"learning_rate": 5.8531067593220333e-05,
"loss": 0.5211,
"step": 137
},
{
"epoch": 1.72,
"learning_rate": 5.830507891525423e-05,
"loss": 0.4851,
"step": 138
},
{
"epoch": 1.73,
"learning_rate": 5.807909023728813e-05,
"loss": 0.5172,
"step": 139
},
{
"epoch": 1.75,
"learning_rate": 5.7853101559322024e-05,
"loss": 0.5038,
"step": 140
},
{
"epoch": 1.75,
"eval_loss": 0.9127333760261536,
"eval_runtime": 22.9168,
"eval_samples_per_second": 10.909,
"eval_steps_per_second": 2.749,
"step": 140
},
{
"epoch": 1.76,
"learning_rate": 5.762711288135593e-05,
"loss": 0.5116,
"step": 141
},
{
"epoch": 1.77,
"learning_rate": 5.740112420338982e-05,
"loss": 0.5009,
"step": 142
},
{
"epoch": 1.78,
"learning_rate": 5.717513552542372e-05,
"loss": 0.4835,
"step": 143
},
{
"epoch": 1.8,
"learning_rate": 5.694914684745762e-05,
"loss": 0.4917,
"step": 144
},
{
"epoch": 1.81,
"learning_rate": 5.672315816949152e-05,
"loss": 0.4873,
"step": 145
},
{
"epoch": 1.82,
"learning_rate": 5.649716949152541e-05,
"loss": 0.5094,
"step": 146
},
{
"epoch": 1.84,
"learning_rate": 5.627118081355932e-05,
"loss": 0.5193,
"step": 147
},
{
"epoch": 1.85,
"learning_rate": 5.604519213559321e-05,
"loss": 0.4921,
"step": 148
},
{
"epoch": 1.86,
"learning_rate": 5.581920345762711e-05,
"loss": 0.4978,
"step": 149
},
{
"epoch": 1.87,
"learning_rate": 5.559321477966101e-05,
"loss": 0.4901,
"step": 150
},
{
"epoch": 1.89,
"learning_rate": 5.536722610169491e-05,
"loss": 0.5042,
"step": 151
},
{
"epoch": 1.9,
"learning_rate": 5.51412374237288e-05,
"loss": 0.4838,
"step": 152
},
{
"epoch": 1.91,
"learning_rate": 5.491524874576271e-05,
"loss": 0.485,
"step": 153
},
{
"epoch": 1.92,
"learning_rate": 5.46892600677966e-05,
"loss": 0.4967,
"step": 154
},
{
"epoch": 1.94,
"learning_rate": 5.44632713898305e-05,
"loss": 0.4943,
"step": 155
},
{
"epoch": 1.95,
"learning_rate": 5.4237282711864406e-05,
"loss": 0.4999,
"step": 156
},
{
"epoch": 1.96,
"learning_rate": 5.40112940338983e-05,
"loss": 0.4456,
"step": 157
},
{
"epoch": 1.97,
"learning_rate": 5.3785305355932205e-05,
"loss": 0.4787,
"step": 158
},
{
"epoch": 1.99,
"learning_rate": 5.35593166779661e-05,
"loss": 0.4594,
"step": 159
},
{
"epoch": 2.0,
"learning_rate": 5.3333327999999996e-05,
"loss": 0.4744,
"step": 160
},
{
"epoch": 2.0,
"eval_loss": 0.9105737805366516,
"eval_runtime": 22.9129,
"eval_samples_per_second": 10.911,
"eval_steps_per_second": 2.75,
"step": 160
},
{
"epoch": 2.01,
"learning_rate": 5.3107339322033896e-05,
"loss": 0.5174,
"step": 161
},
{
"epoch": 2.03,
"learning_rate": 5.2881350644067795e-05,
"loss": 0.46,
"step": 162
}
],
"logging_steps": 1,
"max_steps": 395,
"num_train_epochs": 5,
"save_steps": 500,
"total_flos": 9.165994568020132e+17,
"trial_name": null,
"trial_params": null
}