functionary-medium-v2.2 / trainer_state.json
khaimai's picture
Upload folder using huggingface_hub
e9d6e00
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.998719590268886,
"eval_steps": 75,
"global_step": 390,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 8.333333333333333e-07,
"loss": 2.0203,
"step": 1
},
{
"epoch": 0.01,
"learning_rate": 1.6666666666666667e-06,
"loss": 2.1077,
"step": 2
},
{
"epoch": 0.01,
"learning_rate": 2.5e-06,
"loss": 1.6828,
"step": 3
},
{
"epoch": 0.01,
"learning_rate": 3.3333333333333333e-06,
"loss": 1.7609,
"step": 4
},
{
"epoch": 0.01,
"learning_rate": 4.166666666666667e-06,
"loss": 1.8274,
"step": 5
},
{
"epoch": 0.02,
"learning_rate": 5e-06,
"loss": 1.7816,
"step": 6
},
{
"epoch": 0.02,
"learning_rate": 5.833333333333334e-06,
"loss": 1.592,
"step": 7
},
{
"epoch": 0.02,
"learning_rate": 6.666666666666667e-06,
"loss": 1.4271,
"step": 8
},
{
"epoch": 0.02,
"learning_rate": 7.500000000000001e-06,
"loss": 1.4018,
"step": 9
},
{
"epoch": 0.03,
"learning_rate": 8.333333333333334e-06,
"loss": 1.1137,
"step": 10
},
{
"epoch": 0.03,
"learning_rate": 9.166666666666666e-06,
"loss": 1.2774,
"step": 11
},
{
"epoch": 0.03,
"learning_rate": 1e-05,
"loss": 1.1152,
"step": 12
},
{
"epoch": 0.03,
"learning_rate": 9.999827315381885e-06,
"loss": 1.1456,
"step": 13
},
{
"epoch": 0.04,
"learning_rate": 9.99930927345553e-06,
"loss": 1.0368,
"step": 14
},
{
"epoch": 0.04,
"learning_rate": 9.998445910004082e-06,
"loss": 1.0394,
"step": 15
},
{
"epoch": 0.04,
"learning_rate": 9.99723728466338e-06,
"loss": 1.0852,
"step": 16
},
{
"epoch": 0.04,
"learning_rate": 9.995683480917821e-06,
"loss": 1.0978,
"step": 17
},
{
"epoch": 0.05,
"learning_rate": 9.993784606094612e-06,
"loss": 1.0971,
"step": 18
},
{
"epoch": 0.05,
"learning_rate": 9.991540791356342e-06,
"loss": 0.8742,
"step": 19
},
{
"epoch": 0.05,
"learning_rate": 9.988952191691925e-06,
"loss": 1.0309,
"step": 20
},
{
"epoch": 0.05,
"learning_rate": 9.986018985905901e-06,
"loss": 0.9933,
"step": 21
},
{
"epoch": 0.06,
"learning_rate": 9.982741376606077e-06,
"loss": 1.0179,
"step": 22
},
{
"epoch": 0.06,
"learning_rate": 9.97911959018954e-06,
"loss": 0.9589,
"step": 23
},
{
"epoch": 0.06,
"learning_rate": 9.975153876827008e-06,
"loss": 0.9256,
"step": 24
},
{
"epoch": 0.06,
"learning_rate": 9.97084451044556e-06,
"loss": 0.9052,
"step": 25
},
{
"epoch": 0.07,
"learning_rate": 9.966191788709716e-06,
"loss": 0.7886,
"step": 26
},
{
"epoch": 0.07,
"learning_rate": 9.961196033000862e-06,
"loss": 0.8052,
"step": 27
},
{
"epoch": 0.07,
"learning_rate": 9.955857588395065e-06,
"loss": 0.8644,
"step": 28
},
{
"epoch": 0.07,
"learning_rate": 9.950176823639233e-06,
"loss": 0.943,
"step": 29
},
{
"epoch": 0.08,
"learning_rate": 9.944154131125643e-06,
"loss": 0.8096,
"step": 30
},
{
"epoch": 0.08,
"learning_rate": 9.937789926864838e-06,
"loss": 0.8375,
"step": 31
},
{
"epoch": 0.08,
"learning_rate": 9.931084650456892e-06,
"loss": 0.9596,
"step": 32
},
{
"epoch": 0.08,
"learning_rate": 9.924038765061042e-06,
"loss": 0.8191,
"step": 33
},
{
"epoch": 0.09,
"learning_rate": 9.916652757363698e-06,
"loss": 0.6918,
"step": 34
},
{
"epoch": 0.09,
"learning_rate": 9.90892713754483e-06,
"loss": 0.8118,
"step": 35
},
{
"epoch": 0.09,
"learning_rate": 9.900862439242719e-06,
"loss": 0.8023,
"step": 36
},
{
"epoch": 0.09,
"learning_rate": 9.892459219517108e-06,
"loss": 0.8091,
"step": 37
},
{
"epoch": 0.1,
"learning_rate": 9.883718058810708e-06,
"loss": 0.7172,
"step": 38
},
{
"epoch": 0.1,
"learning_rate": 9.874639560909118e-06,
"loss": 0.8353,
"step": 39
},
{
"epoch": 0.1,
"learning_rate": 9.86522435289912e-06,
"loss": 0.8134,
"step": 40
},
{
"epoch": 0.1,
"learning_rate": 9.855473085125351e-06,
"loss": 0.8074,
"step": 41
},
{
"epoch": 0.11,
"learning_rate": 9.84538643114539e-06,
"loss": 0.6778,
"step": 42
},
{
"epoch": 0.11,
"learning_rate": 9.834965087683237e-06,
"loss": 0.8074,
"step": 43
},
{
"epoch": 0.11,
"learning_rate": 9.824209774581176e-06,
"loss": 0.6441,
"step": 44
},
{
"epoch": 0.12,
"learning_rate": 9.81312123475006e-06,
"loss": 0.6854,
"step": 45
},
{
"epoch": 0.12,
"learning_rate": 9.801700234118e-06,
"loss": 0.7312,
"step": 46
},
{
"epoch": 0.12,
"learning_rate": 9.789947561577445e-06,
"loss": 0.6815,
"step": 47
},
{
"epoch": 0.12,
"learning_rate": 9.777864028930705e-06,
"loss": 0.6503,
"step": 48
},
{
"epoch": 0.13,
"learning_rate": 9.765450470833867e-06,
"loss": 0.6441,
"step": 49
},
{
"epoch": 0.13,
"learning_rate": 9.752707744739146e-06,
"loss": 0.7132,
"step": 50
},
{
"epoch": 0.13,
"learning_rate": 9.73963673083566e-06,
"loss": 0.6414,
"step": 51
},
{
"epoch": 0.13,
"learning_rate": 9.726238331988625e-06,
"loss": 0.6329,
"step": 52
},
{
"epoch": 0.14,
"learning_rate": 9.712513473676997e-06,
"loss": 0.7038,
"step": 53
},
{
"epoch": 0.14,
"learning_rate": 9.698463103929542e-06,
"loss": 0.7058,
"step": 54
},
{
"epoch": 0.14,
"learning_rate": 9.684088193259356e-06,
"loss": 0.7323,
"step": 55
},
{
"epoch": 0.14,
"learning_rate": 9.669389734596819e-06,
"loss": 0.5464,
"step": 56
},
{
"epoch": 0.15,
"learning_rate": 9.654368743221022e-06,
"loss": 0.5878,
"step": 57
},
{
"epoch": 0.15,
"learning_rate": 9.639026256689628e-06,
"loss": 0.4955,
"step": 58
},
{
"epoch": 0.15,
"learning_rate": 9.623363334767208e-06,
"loss": 0.5786,
"step": 59
},
{
"epoch": 0.15,
"learning_rate": 9.60738105935204e-06,
"loss": 0.526,
"step": 60
},
{
"epoch": 0.16,
"learning_rate": 9.591080534401371e-06,
"loss": 0.5261,
"step": 61
},
{
"epoch": 0.16,
"learning_rate": 9.574462885855173e-06,
"loss": 0.6276,
"step": 62
},
{
"epoch": 0.16,
"learning_rate": 9.557529261558367e-06,
"loss": 0.585,
"step": 63
},
{
"epoch": 0.16,
"learning_rate": 9.540280831181525e-06,
"loss": 0.651,
"step": 64
},
{
"epoch": 0.17,
"learning_rate": 9.522718786140096e-06,
"loss": 0.4835,
"step": 65
},
{
"epoch": 0.17,
"learning_rate": 9.504844339512096e-06,
"loss": 0.5757,
"step": 66
},
{
"epoch": 0.17,
"learning_rate": 9.486658725954321e-06,
"loss": 0.6357,
"step": 67
},
{
"epoch": 0.17,
"learning_rate": 9.468163201617063e-06,
"loss": 0.6052,
"step": 68
},
{
"epoch": 0.18,
"learning_rate": 9.449359044057344e-06,
"loss": 0.6328,
"step": 69
},
{
"epoch": 0.18,
"learning_rate": 9.430247552150673e-06,
"loss": 0.4931,
"step": 70
},
{
"epoch": 0.18,
"learning_rate": 9.410830046001321e-06,
"loss": 0.5083,
"step": 71
},
{
"epoch": 0.18,
"learning_rate": 9.391107866851143e-06,
"loss": 0.5612,
"step": 72
},
{
"epoch": 0.19,
"learning_rate": 9.37108237698693e-06,
"loss": 0.5848,
"step": 73
},
{
"epoch": 0.19,
"learning_rate": 9.350754959646306e-06,
"loss": 0.5098,
"step": 74
},
{
"epoch": 0.19,
"learning_rate": 9.330127018922195e-06,
"loss": 0.6561,
"step": 75
},
{
"epoch": 0.19,
"eval_accuracy": 0.8096002455644081,
"eval_accuracy_<|content|>": 0.9054457292055204,
"eval_accuracy_<|from|>": 0.9848293299620733,
"eval_accuracy_<|recipient|>": 0.5600505689001264,
"eval_accuracy_<|stop|>": 0.9188891337888473,
"eval_accuracy_total_num_<|content|>": 5362,
"eval_accuracy_total_num_<|from|>": 791,
"eval_accuracy_total_num_<|recipient|>": 791,
"eval_accuracy_total_num_<|stop|>": 4537,
"eval_loss": NaN,
"eval_perplexity": 1.0685622608271412,
"eval_runtime": 368.5003,
"eval_samples_per_second": 3.731,
"eval_steps_per_second": 0.467,
"step": 75
},
{
"epoch": 0.19,
"learning_rate": 9.309199979665821e-06,
"loss": 0.5904,
"step": 76
},
{
"epoch": 0.2,
"learning_rate": 9.287975287388297e-06,
"loss": 0.5897,
"step": 77
},
{
"epoch": 0.2,
"learning_rate": 9.266454408160779e-06,
"loss": 0.5341,
"step": 78
},
{
"epoch": 0.2,
"learning_rate": 9.244638828513189e-06,
"loss": 0.5282,
"step": 79
},
{
"epoch": 0.2,
"learning_rate": 9.22253005533154e-06,
"loss": 0.4889,
"step": 80
},
{
"epoch": 0.21,
"learning_rate": 9.200129615753858e-06,
"loss": 0.6347,
"step": 81
},
{
"epoch": 0.21,
"learning_rate": 9.177439057064684e-06,
"loss": 0.5477,
"step": 82
},
{
"epoch": 0.21,
"learning_rate": 9.154459946588199e-06,
"loss": 0.5674,
"step": 83
},
{
"epoch": 0.22,
"learning_rate": 9.131193871579975e-06,
"loss": 0.5769,
"step": 84
},
{
"epoch": 0.22,
"learning_rate": 9.107642439117322e-06,
"loss": 0.4182,
"step": 85
},
{
"epoch": 0.22,
"learning_rate": 9.083807275988285e-06,
"loss": 0.6244,
"step": 86
},
{
"epoch": 0.22,
"learning_rate": 9.059690028579285e-06,
"loss": 0.4599,
"step": 87
},
{
"epoch": 0.23,
"learning_rate": 9.035292362761382e-06,
"loss": 0.5209,
"step": 88
},
{
"epoch": 0.23,
"learning_rate": 9.01061596377522e-06,
"loss": 0.5992,
"step": 89
},
{
"epoch": 0.23,
"learning_rate": 8.985662536114614e-06,
"loss": 0.5819,
"step": 90
},
{
"epoch": 0.23,
"learning_rate": 8.960433803408813e-06,
"loss": 0.5248,
"step": 91
},
{
"epoch": 0.24,
"learning_rate": 8.934931508303446e-06,
"loss": 0.497,
"step": 92
},
{
"epoch": 0.24,
"learning_rate": 8.90915741234015e-06,
"loss": 0.4701,
"step": 93
},
{
"epoch": 0.24,
"learning_rate": 8.883113295834893e-06,
"loss": 0.4594,
"step": 94
},
{
"epoch": 0.24,
"learning_rate": 8.856800957755e-06,
"loss": 0.4374,
"step": 95
},
{
"epoch": 0.25,
"learning_rate": 8.83022221559489e-06,
"loss": 0.53,
"step": 96
},
{
"epoch": 0.25,
"learning_rate": 8.803378905250544e-06,
"loss": 0.4277,
"step": 97
},
{
"epoch": 0.25,
"learning_rate": 8.776272880892675e-06,
"loss": 0.4828,
"step": 98
},
{
"epoch": 0.25,
"learning_rate": 8.748906014838672e-06,
"loss": 0.5068,
"step": 99
},
{
"epoch": 0.26,
"learning_rate": 8.721280197423259e-06,
"loss": 0.462,
"step": 100
},
{
"epoch": 0.26,
"learning_rate": 8.69339733686793e-06,
"loss": 0.4336,
"step": 101
},
{
"epoch": 0.26,
"learning_rate": 8.665259359149132e-06,
"loss": 0.4928,
"step": 102
},
{
"epoch": 0.26,
"learning_rate": 8.636868207865244e-06,
"loss": 0.5075,
"step": 103
},
{
"epoch": 0.27,
"learning_rate": 8.608225844102312e-06,
"loss": 0.5317,
"step": 104
},
{
"epoch": 0.27,
"learning_rate": 8.579334246298593e-06,
"loss": 0.5727,
"step": 105
},
{
"epoch": 0.27,
"learning_rate": 8.550195410107903e-06,
"loss": 0.5007,
"step": 106
},
{
"epoch": 0.27,
"learning_rate": 8.52081134826176e-06,
"loss": 0.5853,
"step": 107
},
{
"epoch": 0.28,
"learning_rate": 8.491184090430365e-06,
"loss": 0.5461,
"step": 108
},
{
"epoch": 0.28,
"learning_rate": 8.461315683082398e-06,
"loss": 0.5821,
"step": 109
},
{
"epoch": 0.28,
"learning_rate": 8.43120818934367e-06,
"loss": 0.5009,
"step": 110
},
{
"epoch": 0.28,
"learning_rate": 8.400863688854598e-06,
"loss": 0.4381,
"step": 111
},
{
"epoch": 0.29,
"learning_rate": 8.370284277626576e-06,
"loss": 0.4998,
"step": 112
},
{
"epoch": 0.29,
"learning_rate": 8.339472067897187e-06,
"loss": 0.554,
"step": 113
},
{
"epoch": 0.29,
"learning_rate": 8.308429187984298e-06,
"loss": 0.4264,
"step": 114
},
{
"epoch": 0.29,
"learning_rate": 8.277157782139051e-06,
"loss": 0.4692,
"step": 115
},
{
"epoch": 0.3,
"learning_rate": 8.24566001039776e-06,
"loss": 0.5083,
"step": 116
},
{
"epoch": 0.3,
"learning_rate": 8.213938048432697e-06,
"loss": 0.5876,
"step": 117
},
{
"epoch": 0.3,
"learning_rate": 8.181994087401819e-06,
"loss": 0.4498,
"step": 118
},
{
"epoch": 0.3,
"learning_rate": 8.149830333797407e-06,
"loss": 0.5222,
"step": 119
},
{
"epoch": 0.31,
"learning_rate": 8.117449009293668e-06,
"loss": 0.5295,
"step": 120
},
{
"epoch": 0.31,
"learning_rate": 8.084852350593264e-06,
"loss": 0.492,
"step": 121
},
{
"epoch": 0.31,
"learning_rate": 8.052042609272817e-06,
"loss": 0.675,
"step": 122
},
{
"epoch": 0.31,
"learning_rate": 8.019022051627387e-06,
"loss": 0.4563,
"step": 123
},
{
"epoch": 0.32,
"learning_rate": 7.985792958513932e-06,
"loss": 0.5108,
"step": 124
},
{
"epoch": 0.32,
"learning_rate": 7.952357625193749e-06,
"loss": 0.4258,
"step": 125
},
{
"epoch": 0.32,
"learning_rate": 7.918718361173951e-06,
"loss": 0.4717,
"step": 126
},
{
"epoch": 0.33,
"learning_rate": 7.884877490047915e-06,
"loss": 0.529,
"step": 127
},
{
"epoch": 0.33,
"learning_rate": 7.85083734933481e-06,
"loss": 0.4353,
"step": 128
},
{
"epoch": 0.33,
"learning_rate": 7.81660029031811e-06,
"loss": 0.512,
"step": 129
},
{
"epoch": 0.33,
"learning_rate": 7.782168677883206e-06,
"loss": 0.5723,
"step": 130
},
{
"epoch": 0.34,
"learning_rate": 7.747544890354031e-06,
"loss": 0.5354,
"step": 131
},
{
"epoch": 0.34,
"learning_rate": 7.712731319328798e-06,
"loss": 0.4359,
"step": 132
},
{
"epoch": 0.34,
"learning_rate": 7.677730369514792e-06,
"loss": 0.3552,
"step": 133
},
{
"epoch": 0.34,
"learning_rate": 7.642544458562278e-06,
"loss": 0.4758,
"step": 134
},
{
"epoch": 0.35,
"learning_rate": 7.607176016897491e-06,
"loss": 0.4783,
"step": 135
},
{
"epoch": 0.35,
"learning_rate": 7.571627487554769e-06,
"loss": 0.6152,
"step": 136
},
{
"epoch": 0.35,
"learning_rate": 7.535901326007796e-06,
"loss": 0.4984,
"step": 137
},
{
"epoch": 0.35,
"learning_rate": 7.500000000000001e-06,
"loss": 0.4241,
"step": 138
},
{
"epoch": 0.36,
"learning_rate": 7.463925989374089e-06,
"loss": 0.4422,
"step": 139
},
{
"epoch": 0.36,
"learning_rate": 7.4276817859007615e-06,
"loss": 0.475,
"step": 140
},
{
"epoch": 0.36,
"learning_rate": 7.391269893106592e-06,
"loss": 0.4778,
"step": 141
},
{
"epoch": 0.36,
"learning_rate": 7.354692826101102e-06,
"loss": 0.406,
"step": 142
},
{
"epoch": 0.37,
"learning_rate": 7.317953111403029e-06,
"loss": 0.499,
"step": 143
},
{
"epoch": 0.37,
"learning_rate": 7.281053286765816e-06,
"loss": 0.6181,
"step": 144
},
{
"epoch": 0.37,
"learning_rate": 7.243995901002312e-06,
"loss": 0.6451,
"step": 145
},
{
"epoch": 0.37,
"learning_rate": 7.206783513808721e-06,
"loss": 0.4481,
"step": 146
},
{
"epoch": 0.38,
"learning_rate": 7.169418695587791e-06,
"loss": 0.4808,
"step": 147
},
{
"epoch": 0.38,
"learning_rate": 7.1319040272712705e-06,
"loss": 0.5812,
"step": 148
},
{
"epoch": 0.38,
"learning_rate": 7.094242100141625e-06,
"loss": 0.5282,
"step": 149
},
{
"epoch": 0.38,
"learning_rate": 7.056435515653059e-06,
"loss": 0.508,
"step": 150
},
{
"epoch": 0.38,
"eval_accuracy": 0.8220327916670906,
"eval_accuracy_<|content|>": 0.9985080193957478,
"eval_accuracy_<|from|>": 0.97724399494311,
"eval_accuracy_<|recipient|>": 1.0,
"eval_accuracy_<|stop|>": 0.84174564690324,
"eval_accuracy_total_num_<|content|>": 5362,
"eval_accuracy_total_num_<|from|>": 791,
"eval_accuracy_total_num_<|recipient|>": 791,
"eval_accuracy_total_num_<|stop|>": 4537,
"eval_loss": NaN,
"eval_perplexity": 1.0626254108805258,
"eval_runtime": 331.6718,
"eval_samples_per_second": 4.146,
"eval_steps_per_second": 0.519,
"step": 150
},
{
"epoch": 0.39,
"learning_rate": 7.0184868852518114e-06,
"loss": 0.3785,
"step": 151
},
{
"epoch": 0.39,
"learning_rate": 6.980398830195785e-06,
"loss": 0.3625,
"step": 152
},
{
"epoch": 0.39,
"learning_rate": 6.942173981373474e-06,
"loss": 0.5124,
"step": 153
},
{
"epoch": 0.39,
"learning_rate": 6.903814979122249e-06,
"loss": 0.5874,
"step": 154
},
{
"epoch": 0.4,
"learning_rate": 6.86532447304597e-06,
"loss": 0.473,
"step": 155
},
{
"epoch": 0.4,
"learning_rate": 6.8267051218319766e-06,
"loss": 0.5132,
"step": 156
},
{
"epoch": 0.4,
"learning_rate": 6.787959593067431e-06,
"loss": 0.597,
"step": 157
},
{
"epoch": 0.4,
"learning_rate": 6.749090563055075e-06,
"loss": 0.4303,
"step": 158
},
{
"epoch": 0.41,
"learning_rate": 6.710100716628345e-06,
"loss": 0.5029,
"step": 159
},
{
"epoch": 0.41,
"learning_rate": 6.6709927469659385e-06,
"loss": 0.5159,
"step": 160
},
{
"epoch": 0.41,
"learning_rate": 6.631769355405779e-06,
"loss": 0.4789,
"step": 161
},
{
"epoch": 0.41,
"learning_rate": 6.592433251258423e-06,
"loss": 0.4632,
"step": 162
},
{
"epoch": 0.42,
"learning_rate": 6.552987151619919e-06,
"loss": 0.5442,
"step": 163
},
{
"epoch": 0.42,
"learning_rate": 6.513433781184131e-06,
"loss": 0.4984,
"step": 164
},
{
"epoch": 0.42,
"learning_rate": 6.473775872054522e-06,
"loss": 0.4579,
"step": 165
},
{
"epoch": 0.43,
"learning_rate": 6.434016163555452e-06,
"loss": 0.434,
"step": 166
},
{
"epoch": 0.43,
"learning_rate": 6.394157402042952e-06,
"loss": 0.5519,
"step": 167
},
{
"epoch": 0.43,
"learning_rate": 6.354202340715027e-06,
"loss": 0.4145,
"step": 168
},
{
"epoch": 0.43,
"learning_rate": 6.314153739421477e-06,
"loss": 0.4358,
"step": 169
},
{
"epoch": 0.44,
"learning_rate": 6.274014364473274e-06,
"loss": 0.4972,
"step": 170
},
{
"epoch": 0.44,
"learning_rate": 6.233786988451468e-06,
"loss": 0.5023,
"step": 171
},
{
"epoch": 0.44,
"learning_rate": 6.19347439001569e-06,
"loss": 0.4899,
"step": 172
},
{
"epoch": 0.44,
"learning_rate": 6.153079353712201e-06,
"loss": 0.4607,
"step": 173
},
{
"epoch": 0.45,
"learning_rate": 6.112604669781572e-06,
"loss": 0.4166,
"step": 174
},
{
"epoch": 0.45,
"learning_rate": 6.0720531339659386e-06,
"loss": 0.4975,
"step": 175
},
{
"epoch": 0.45,
"learning_rate": 6.031427547315889e-06,
"loss": 0.4031,
"step": 176
},
{
"epoch": 0.45,
"learning_rate": 5.990730715996989e-06,
"loss": 0.5201,
"step": 177
},
{
"epoch": 0.46,
"learning_rate": 5.949965451095952e-06,
"loss": 0.4893,
"step": 178
},
{
"epoch": 0.46,
"learning_rate": 5.909134568426455e-06,
"loss": 0.61,
"step": 179
},
{
"epoch": 0.46,
"learning_rate": 5.8682408883346535e-06,
"loss": 0.4173,
"step": 180
},
{
"epoch": 0.46,
"learning_rate": 5.827287235504356e-06,
"loss": 0.4462,
"step": 181
},
{
"epoch": 0.47,
"learning_rate": 5.786276438761928e-06,
"loss": 0.53,
"step": 182
},
{
"epoch": 0.47,
"learning_rate": 5.745211330880872e-06,
"loss": 0.3297,
"step": 183
},
{
"epoch": 0.47,
"learning_rate": 5.7040947483861845e-06,
"loss": 0.466,
"step": 184
},
{
"epoch": 0.47,
"learning_rate": 5.6629295313583975e-06,
"loss": 0.5355,
"step": 185
},
{
"epoch": 0.48,
"learning_rate": 5.621718523237427e-06,
"loss": 0.4806,
"step": 186
},
{
"epoch": 0.48,
"learning_rate": 5.5804645706261515e-06,
"loss": 0.3791,
"step": 187
},
{
"epoch": 0.48,
"learning_rate": 5.539170523093794e-06,
"loss": 0.5198,
"step": 188
},
{
"epoch": 0.48,
"learning_rate": 5.497839232979084e-06,
"loss": 0.4091,
"step": 189
},
{
"epoch": 0.49,
"learning_rate": 5.456473555193242e-06,
"loss": 0.3974,
"step": 190
},
{
"epoch": 0.49,
"learning_rate": 5.415076347022777e-06,
"loss": 0.4699,
"step": 191
},
{
"epoch": 0.49,
"learning_rate": 5.373650467932122e-06,
"loss": 0.4385,
"step": 192
},
{
"epoch": 0.49,
"learning_rate": 5.332198779366123e-06,
"loss": 0.4943,
"step": 193
},
{
"epoch": 0.5,
"learning_rate": 5.290724144552379e-06,
"loss": 0.4705,
"step": 194
},
{
"epoch": 0.5,
"learning_rate": 5.249229428303486e-06,
"loss": 0.5045,
"step": 195
},
{
"epoch": 0.5,
"learning_rate": 5.207717496819134e-06,
"loss": 0.5795,
"step": 196
},
{
"epoch": 0.5,
"learning_rate": 5.166191217488134e-06,
"loss": 0.4442,
"step": 197
},
{
"epoch": 0.51,
"learning_rate": 5.1246534586903655e-06,
"loss": 0.4642,
"step": 198
},
{
"epoch": 0.51,
"learning_rate": 5.083107089598632e-06,
"loss": 0.4908,
"step": 199
},
{
"epoch": 0.51,
"learning_rate": 5.041554979980487e-06,
"loss": 0.4577,
"step": 200
},
{
"epoch": 0.51,
"learning_rate": 5e-06,
"loss": 0.402,
"step": 201
},
{
"epoch": 0.52,
"learning_rate": 4.958445020019516e-06,
"loss": 0.4191,
"step": 202
},
{
"epoch": 0.52,
"learning_rate": 4.916892910401369e-06,
"loss": 0.4828,
"step": 203
},
{
"epoch": 0.52,
"learning_rate": 4.875346541309637e-06,
"loss": 0.4478,
"step": 204
},
{
"epoch": 0.52,
"learning_rate": 4.833808782511867e-06,
"loss": 0.5202,
"step": 205
},
{
"epoch": 0.53,
"learning_rate": 4.792282503180867e-06,
"loss": 0.3495,
"step": 206
},
{
"epoch": 0.53,
"learning_rate": 4.750770571696514e-06,
"loss": 0.5703,
"step": 207
},
{
"epoch": 0.53,
"learning_rate": 4.7092758554476215e-06,
"loss": 0.3986,
"step": 208
},
{
"epoch": 0.54,
"learning_rate": 4.66780122063388e-06,
"loss": 0.3185,
"step": 209
},
{
"epoch": 0.54,
"learning_rate": 4.626349532067879e-06,
"loss": 0.4846,
"step": 210
},
{
"epoch": 0.54,
"learning_rate": 4.584923652977224e-06,
"loss": 0.4592,
"step": 211
},
{
"epoch": 0.54,
"learning_rate": 4.5435264448067595e-06,
"loss": 0.4666,
"step": 212
},
{
"epoch": 0.55,
"learning_rate": 4.502160767020918e-06,
"loss": 0.428,
"step": 213
},
{
"epoch": 0.55,
"learning_rate": 4.460829476906208e-06,
"loss": 0.5192,
"step": 214
},
{
"epoch": 0.55,
"learning_rate": 4.4195354293738484e-06,
"loss": 0.4456,
"step": 215
},
{
"epoch": 0.55,
"learning_rate": 4.3782814767625755e-06,
"loss": 0.5031,
"step": 216
},
{
"epoch": 0.56,
"learning_rate": 4.337070468641604e-06,
"loss": 0.3835,
"step": 217
},
{
"epoch": 0.56,
"learning_rate": 4.295905251613817e-06,
"loss": 0.4406,
"step": 218
},
{
"epoch": 0.56,
"learning_rate": 4.254788669119127e-06,
"loss": 0.4374,
"step": 219
},
{
"epoch": 0.56,
"learning_rate": 4.213723561238074e-06,
"loss": 0.4021,
"step": 220
},
{
"epoch": 0.57,
"learning_rate": 4.172712764495645e-06,
"loss": 0.5134,
"step": 221
},
{
"epoch": 0.57,
"learning_rate": 4.131759111665349e-06,
"loss": 0.4838,
"step": 222
},
{
"epoch": 0.57,
"learning_rate": 4.090865431573547e-06,
"loss": 0.4011,
"step": 223
},
{
"epoch": 0.57,
"learning_rate": 4.0500345489040515e-06,
"loss": 0.3548,
"step": 224
},
{
"epoch": 0.58,
"learning_rate": 4.009269284003014e-06,
"loss": 0.391,
"step": 225
},
{
"epoch": 0.58,
"eval_accuracy": 0.8278573148685179,
"eval_accuracy_<|content|>": 0.9986945169712794,
"eval_accuracy_<|from|>": 0.9949431099873578,
"eval_accuracy_<|recipient|>": 1.0,
"eval_accuracy_<|stop|>": 0.939607670266696,
"eval_accuracy_total_num_<|content|>": 5362,
"eval_accuracy_total_num_<|from|>": 791,
"eval_accuracy_total_num_<|recipient|>": 791,
"eval_accuracy_total_num_<|stop|>": 4537,
"eval_loss": NaN,
"eval_perplexity": 1.0600965829681877,
"eval_runtime": 334.7156,
"eval_samples_per_second": 4.108,
"eval_steps_per_second": 0.514,
"step": 225
},
{
"epoch": 0.58,
"learning_rate": 3.968572452684113e-06,
"loss": 0.5301,
"step": 226
},
{
"epoch": 0.58,
"learning_rate": 3.927946866034062e-06,
"loss": 0.4173,
"step": 227
},
{
"epoch": 0.58,
"learning_rate": 3.887395330218429e-06,
"loss": 0.5295,
"step": 228
},
{
"epoch": 0.59,
"learning_rate": 3.8469206462878e-06,
"loss": 0.3869,
"step": 229
},
{
"epoch": 0.59,
"learning_rate": 3.806525609984312e-06,
"loss": 0.4667,
"step": 230
},
{
"epoch": 0.59,
"learning_rate": 3.7662130115485317e-06,
"loss": 0.4839,
"step": 231
},
{
"epoch": 0.59,
"learning_rate": 3.7259856355267275e-06,
"loss": 0.4778,
"step": 232
},
{
"epoch": 0.6,
"learning_rate": 3.685846260578524e-06,
"loss": 0.4695,
"step": 233
},
{
"epoch": 0.6,
"learning_rate": 3.6457976592849753e-06,
"loss": 0.4,
"step": 234
},
{
"epoch": 0.6,
"learning_rate": 3.6058425979570482e-06,
"loss": 0.441,
"step": 235
},
{
"epoch": 0.6,
"learning_rate": 3.5659838364445505e-06,
"loss": 0.4077,
"step": 236
},
{
"epoch": 0.61,
"learning_rate": 3.526224127945479e-06,
"loss": 0.4368,
"step": 237
},
{
"epoch": 0.61,
"learning_rate": 3.4865662188158713e-06,
"loss": 0.5274,
"step": 238
},
{
"epoch": 0.61,
"learning_rate": 3.4470128483800813e-06,
"loss": 0.4905,
"step": 239
},
{
"epoch": 0.61,
"learning_rate": 3.4075667487415785e-06,
"loss": 0.5323,
"step": 240
},
{
"epoch": 0.62,
"learning_rate": 3.3682306445942224e-06,
"loss": 0.5451,
"step": 241
},
{
"epoch": 0.62,
"learning_rate": 3.3290072530340628e-06,
"loss": 0.4653,
"step": 242
},
{
"epoch": 0.62,
"learning_rate": 3.289899283371657e-06,
"loss": 0.4957,
"step": 243
},
{
"epoch": 0.62,
"learning_rate": 3.250909436944928e-06,
"loss": 0.423,
"step": 244
},
{
"epoch": 0.63,
"learning_rate": 3.2120404069325695e-06,
"loss": 0.5153,
"step": 245
},
{
"epoch": 0.63,
"learning_rate": 3.173294878168025e-06,
"loss": 0.4329,
"step": 246
},
{
"epoch": 0.63,
"learning_rate": 3.1346755269540303e-06,
"loss": 0.377,
"step": 247
},
{
"epoch": 0.64,
"learning_rate": 3.0961850208777527e-06,
"loss": 0.4111,
"step": 248
},
{
"epoch": 0.64,
"learning_rate": 3.057826018626527e-06,
"loss": 0.3951,
"step": 249
},
{
"epoch": 0.64,
"learning_rate": 3.019601169804216e-06,
"loss": 0.5059,
"step": 250
},
{
"epoch": 0.64,
"learning_rate": 2.981513114748189e-06,
"loss": 0.4764,
"step": 251
},
{
"epoch": 0.65,
"learning_rate": 2.9435644843469434e-06,
"loss": 0.4289,
"step": 252
},
{
"epoch": 0.65,
"learning_rate": 2.905757899858377e-06,
"loss": 0.26,
"step": 253
},
{
"epoch": 0.65,
"learning_rate": 2.8680959727287316e-06,
"loss": 0.4835,
"step": 254
},
{
"epoch": 0.65,
"learning_rate": 2.83058130441221e-06,
"loss": 0.4426,
"step": 255
},
{
"epoch": 0.66,
"learning_rate": 2.7932164861912805e-06,
"loss": 0.4571,
"step": 256
},
{
"epoch": 0.66,
"learning_rate": 2.7560040989976894e-06,
"loss": 0.5578,
"step": 257
},
{
"epoch": 0.66,
"learning_rate": 2.718946713234185e-06,
"loss": 0.3772,
"step": 258
},
{
"epoch": 0.66,
"learning_rate": 2.682046888596972e-06,
"loss": 0.4563,
"step": 259
},
{
"epoch": 0.67,
"learning_rate": 2.645307173898901e-06,
"loss": 0.5051,
"step": 260
},
{
"epoch": 0.67,
"learning_rate": 2.608730106893411e-06,
"loss": 0.4382,
"step": 261
},
{
"epoch": 0.67,
"learning_rate": 2.5723182140992385e-06,
"loss": 0.4664,
"step": 262
},
{
"epoch": 0.67,
"learning_rate": 2.536074010625911e-06,
"loss": 0.3587,
"step": 263
},
{
"epoch": 0.68,
"learning_rate": 2.5000000000000015e-06,
"loss": 0.3431,
"step": 264
},
{
"epoch": 0.68,
"learning_rate": 2.464098673992205e-06,
"loss": 0.3606,
"step": 265
},
{
"epoch": 0.68,
"learning_rate": 2.428372512445233e-06,
"loss": 0.4422,
"step": 266
},
{
"epoch": 0.68,
"learning_rate": 2.39282398310251e-06,
"loss": 0.5567,
"step": 267
},
{
"epoch": 0.69,
"learning_rate": 2.357455541437723e-06,
"loss": 0.4313,
"step": 268
},
{
"epoch": 0.69,
"learning_rate": 2.3222696304852084e-06,
"loss": 0.5045,
"step": 269
},
{
"epoch": 0.69,
"learning_rate": 2.2872686806712037e-06,
"loss": 0.4603,
"step": 270
},
{
"epoch": 0.69,
"learning_rate": 2.2524551096459703e-06,
"loss": 0.5706,
"step": 271
},
{
"epoch": 0.7,
"learning_rate": 2.217831322116797e-06,
"loss": 0.4547,
"step": 272
},
{
"epoch": 0.7,
"learning_rate": 2.1833997096818897e-06,
"loss": 0.4556,
"step": 273
},
{
"epoch": 0.7,
"learning_rate": 2.1491626506651914e-06,
"loss": 0.5204,
"step": 274
},
{
"epoch": 0.7,
"learning_rate": 2.115122509952085e-06,
"loss": 0.3224,
"step": 275
},
{
"epoch": 0.71,
"learning_rate": 2.081281638826052e-06,
"loss": 0.5885,
"step": 276
},
{
"epoch": 0.71,
"learning_rate": 2.047642374806252e-06,
"loss": 0.6023,
"step": 277
},
{
"epoch": 0.71,
"learning_rate": 2.0142070414860704e-06,
"loss": 0.4437,
"step": 278
},
{
"epoch": 0.71,
"learning_rate": 1.980977948372612e-06,
"loss": 0.4753,
"step": 279
},
{
"epoch": 0.72,
"learning_rate": 1.947957390727185e-06,
"loss": 0.4953,
"step": 280
},
{
"epoch": 0.72,
"learning_rate": 1.9151476494067376e-06,
"loss": 0.379,
"step": 281
},
{
"epoch": 0.72,
"learning_rate": 1.8825509907063328e-06,
"loss": 0.459,
"step": 282
},
{
"epoch": 0.72,
"learning_rate": 1.8501696662025937e-06,
"loss": 0.5653,
"step": 283
},
{
"epoch": 0.73,
"learning_rate": 1.8180059125981826e-06,
"loss": 0.3608,
"step": 284
},
{
"epoch": 0.73,
"learning_rate": 1.7860619515673034e-06,
"loss": 0.4926,
"step": 285
},
{
"epoch": 0.73,
"learning_rate": 1.7543399896022406e-06,
"loss": 0.3828,
"step": 286
},
{
"epoch": 0.73,
"learning_rate": 1.7228422178609488e-06,
"loss": 0.3937,
"step": 287
},
{
"epoch": 0.74,
"learning_rate": 1.6915708120157042e-06,
"loss": 0.3875,
"step": 288
},
{
"epoch": 0.74,
"learning_rate": 1.6605279321028138e-06,
"loss": 0.4678,
"step": 289
},
{
"epoch": 0.74,
"learning_rate": 1.6297157223734228e-06,
"loss": 0.5462,
"step": 290
},
{
"epoch": 0.75,
"learning_rate": 1.5991363111454023e-06,
"loss": 0.476,
"step": 291
},
{
"epoch": 0.75,
"learning_rate": 1.5687918106563326e-06,
"loss": 0.5062,
"step": 292
},
{
"epoch": 0.75,
"learning_rate": 1.5386843169176025e-06,
"loss": 0.5062,
"step": 293
},
{
"epoch": 0.75,
"learning_rate": 1.5088159095696365e-06,
"loss": 0.4348,
"step": 294
},
{
"epoch": 0.76,
"learning_rate": 1.4791886517382415e-06,
"loss": 0.3801,
"step": 295
},
{
"epoch": 0.76,
"learning_rate": 1.4498045898920988e-06,
"loss": 0.4874,
"step": 296
},
{
"epoch": 0.76,
"learning_rate": 1.4206657537014078e-06,
"loss": 0.5251,
"step": 297
},
{
"epoch": 0.76,
"learning_rate": 1.3917741558976894e-06,
"loss": 0.3894,
"step": 298
},
{
"epoch": 0.77,
"learning_rate": 1.3631317921347564e-06,
"loss": 0.3325,
"step": 299
},
{
"epoch": 0.77,
"learning_rate": 1.3347406408508695e-06,
"loss": 0.4508,
"step": 300
},
{
"epoch": 0.77,
"eval_accuracy": 0.8337530653153841,
"eval_accuracy_<|content|>": 1.0,
"eval_accuracy_<|from|>": 0.9911504424778761,
"eval_accuracy_<|recipient|>": 1.0,
"eval_accuracy_<|stop|>": 0.9329953713907868,
"eval_accuracy_total_num_<|content|>": 5362,
"eval_accuracy_total_num_<|from|>": 791,
"eval_accuracy_total_num_<|recipient|>": 791,
"eval_accuracy_total_num_<|stop|>": 4537,
"eval_loss": NaN,
"eval_perplexity": 1.0576072932064766,
"eval_runtime": 333.0641,
"eval_samples_per_second": 4.128,
"eval_steps_per_second": 0.516,
"step": 300
},
{
"epoch": 0.77,
"learning_rate": 1.3066026631320733e-06,
"loss": 0.3393,
"step": 301
},
{
"epoch": 0.77,
"learning_rate": 1.2787198025767417e-06,
"loss": 0.5482,
"step": 302
},
{
"epoch": 0.78,
"learning_rate": 1.2510939851613285e-06,
"loss": 0.4206,
"step": 303
},
{
"epoch": 0.78,
"learning_rate": 1.223727119107327e-06,
"loss": 0.4613,
"step": 304
},
{
"epoch": 0.78,
"learning_rate": 1.1966210947494583e-06,
"loss": 0.3015,
"step": 305
},
{
"epoch": 0.78,
"learning_rate": 1.1697777844051105e-06,
"loss": 0.4651,
"step": 306
},
{
"epoch": 0.79,
"learning_rate": 1.1431990422450018e-06,
"loss": 0.3467,
"step": 307
},
{
"epoch": 0.79,
"learning_rate": 1.1168867041651082e-06,
"loss": 0.4167,
"step": 308
},
{
"epoch": 0.79,
"learning_rate": 1.0908425876598512e-06,
"loss": 0.4565,
"step": 309
},
{
"epoch": 0.79,
"learning_rate": 1.065068491696556e-06,
"loss": 0.5543,
"step": 310
},
{
"epoch": 0.8,
"learning_rate": 1.0395661965911891e-06,
"loss": 0.4687,
"step": 311
},
{
"epoch": 0.8,
"learning_rate": 1.0143374638853892e-06,
"loss": 0.4595,
"step": 312
},
{
"epoch": 0.8,
"learning_rate": 9.893840362247809e-07,
"loss": 0.4511,
"step": 313
},
{
"epoch": 0.8,
"learning_rate": 9.647076372386195e-07,
"loss": 0.4676,
"step": 314
},
{
"epoch": 0.81,
"learning_rate": 9.403099714207175e-07,
"loss": 0.4385,
"step": 315
},
{
"epoch": 0.81,
"learning_rate": 9.161927240117174e-07,
"loss": 0.3807,
"step": 316
},
{
"epoch": 0.81,
"learning_rate": 8.923575608826812e-07,
"loss": 0.4795,
"step": 317
},
{
"epoch": 0.81,
"learning_rate": 8.688061284200266e-07,
"loss": 0.5137,
"step": 318
},
{
"epoch": 0.82,
"learning_rate": 8.455400534118008e-07,
"loss": 0.4358,
"step": 319
},
{
"epoch": 0.82,
"learning_rate": 8.225609429353187e-07,
"loss": 0.5148,
"step": 320
},
{
"epoch": 0.82,
"learning_rate": 7.99870384246143e-07,
"loss": 0.4713,
"step": 321
},
{
"epoch": 0.82,
"learning_rate": 7.774699446684608e-07,
"loss": 0.4893,
"step": 322
},
{
"epoch": 0.83,
"learning_rate": 7.553611714868136e-07,
"loss": 0.4368,
"step": 323
},
{
"epoch": 0.83,
"learning_rate": 7.33545591839222e-07,
"loss": 0.4275,
"step": 324
},
{
"epoch": 0.83,
"learning_rate": 7.120247126117025e-07,
"loss": 0.4828,
"step": 325
},
{
"epoch": 0.83,
"learning_rate": 6.908000203341802e-07,
"loss": 0.5223,
"step": 326
},
{
"epoch": 0.84,
"learning_rate": 6.698729810778065e-07,
"loss": 0.4837,
"step": 327
},
{
"epoch": 0.84,
"learning_rate": 6.492450403536959e-07,
"loss": 0.4888,
"step": 328
},
{
"epoch": 0.84,
"learning_rate": 6.289176230130728e-07,
"loss": 0.4453,
"step": 329
},
{
"epoch": 0.85,
"learning_rate": 6.088921331488568e-07,
"loss": 0.4143,
"step": 330
},
{
"epoch": 0.85,
"learning_rate": 5.891699539986789e-07,
"loss": 0.4453,
"step": 331
},
{
"epoch": 0.85,
"learning_rate": 5.697524478493288e-07,
"loss": 0.4139,
"step": 332
},
{
"epoch": 0.85,
"learning_rate": 5.506409559426573e-07,
"loss": 0.4899,
"step": 333
},
{
"epoch": 0.86,
"learning_rate": 5.318367983829393e-07,
"loss": 0.3488,
"step": 334
},
{
"epoch": 0.86,
"learning_rate": 5.133412740456805e-07,
"loss": 0.5504,
"step": 335
},
{
"epoch": 0.86,
"learning_rate": 4.951556604879049e-07,
"loss": 0.2987,
"step": 336
},
{
"epoch": 0.86,
"learning_rate": 4.772812138599043e-07,
"loss": 0.503,
"step": 337
},
{
"epoch": 0.87,
"learning_rate": 4.5971916881847543e-07,
"loss": 0.3901,
"step": 338
},
{
"epoch": 0.87,
"learning_rate": 4.4247073844163434e-07,
"loss": 0.4166,
"step": 339
},
{
"epoch": 0.87,
"learning_rate": 4.255371141448272e-07,
"loss": 0.4068,
"step": 340
},
{
"epoch": 0.87,
"learning_rate": 4.089194655986306e-07,
"loss": 0.4019,
"step": 341
},
{
"epoch": 0.88,
"learning_rate": 3.9261894064796136e-07,
"loss": 0.4503,
"step": 342
},
{
"epoch": 0.88,
"learning_rate": 3.766366652327924e-07,
"loss": 0.4989,
"step": 343
},
{
"epoch": 0.88,
"learning_rate": 3.6097374331037326e-07,
"loss": 0.4407,
"step": 344
},
{
"epoch": 0.88,
"learning_rate": 3.4563125677897936e-07,
"loss": 0.3357,
"step": 345
},
{
"epoch": 0.89,
"learning_rate": 3.306102654031823e-07,
"loss": 0.4232,
"step": 346
},
{
"epoch": 0.89,
"learning_rate": 3.1591180674064584e-07,
"loss": 0.3591,
"step": 347
},
{
"epoch": 0.89,
"learning_rate": 3.015368960704584e-07,
"loss": 0.4366,
"step": 348
},
{
"epoch": 0.89,
"learning_rate": 2.8748652632300367e-07,
"loss": 0.4547,
"step": 349
},
{
"epoch": 0.9,
"learning_rate": 2.737616680113758e-07,
"loss": 0.386,
"step": 350
},
{
"epoch": 0.9,
"learning_rate": 2.6036326916434153e-07,
"loss": 0.4585,
"step": 351
},
{
"epoch": 0.9,
"learning_rate": 2.472922552608559e-07,
"loss": 0.4554,
"step": 352
},
{
"epoch": 0.9,
"learning_rate": 2.3454952916613482e-07,
"loss": 0.4524,
"step": 353
},
{
"epoch": 0.91,
"learning_rate": 2.2213597106929608e-07,
"loss": 0.3612,
"step": 354
},
{
"epoch": 0.91,
"learning_rate": 2.1005243842255552e-07,
"loss": 0.4398,
"step": 355
},
{
"epoch": 0.91,
"learning_rate": 1.982997658820013e-07,
"loss": 0.3687,
"step": 356
},
{
"epoch": 0.91,
"learning_rate": 1.8687876524993987e-07,
"loss": 0.3961,
"step": 357
},
{
"epoch": 0.92,
"learning_rate": 1.757902254188254e-07,
"loss": 0.4561,
"step": 358
},
{
"epoch": 0.92,
"learning_rate": 1.6503491231676382e-07,
"loss": 0.4427,
"step": 359
},
{
"epoch": 0.92,
"learning_rate": 1.5461356885461077e-07,
"loss": 0.4432,
"step": 360
},
{
"epoch": 0.92,
"learning_rate": 1.4452691487465087e-07,
"loss": 0.4639,
"step": 361
},
{
"epoch": 0.93,
"learning_rate": 1.3477564710088097e-07,
"loss": 0.4095,
"step": 362
},
{
"epoch": 0.93,
"learning_rate": 1.253604390908819e-07,
"loss": 0.3953,
"step": 363
},
{
"epoch": 0.93,
"learning_rate": 1.1628194118929403e-07,
"loss": 0.4071,
"step": 364
},
{
"epoch": 0.93,
"learning_rate": 1.0754078048289374e-07,
"loss": 0.4452,
"step": 365
},
{
"epoch": 0.94,
"learning_rate": 9.913756075728088e-08,
"loss": 0.433,
"step": 366
},
{
"epoch": 0.94,
"learning_rate": 9.1072862455171e-08,
"loss": 0.4269,
"step": 367
},
{
"epoch": 0.94,
"learning_rate": 8.334724263630301e-08,
"loss": 0.4502,
"step": 368
},
{
"epoch": 0.94,
"learning_rate": 7.59612349389599e-08,
"loss": 0.3055,
"step": 369
},
{
"epoch": 0.95,
"learning_rate": 6.891534954310886e-08,
"loss": 0.4559,
"step": 370
},
{
"epoch": 0.95,
"learning_rate": 6.221007313516159e-08,
"loss": 0.3325,
"step": 371
},
{
"epoch": 0.95,
"learning_rate": 5.584586887435739e-08,
"loss": 0.5016,
"step": 372
},
{
"epoch": 0.96,
"learning_rate": 4.9823176360768166e-08,
"loss": 0.363,
"step": 373
},
{
"epoch": 0.96,
"learning_rate": 4.41424116049366e-08,
"loss": 0.5222,
"step": 374
},
{
"epoch": 0.96,
"learning_rate": 3.8803966999139686e-08,
"loss": 0.4091,
"step": 375
},
{
"epoch": 0.96,
"eval_accuracy": 0.8351479322052294,
"eval_accuracy_<|content|>": 1.0,
"eval_accuracy_<|from|>": 0.9924146649810367,
"eval_accuracy_<|recipient|>": 1.0,
"eval_accuracy_<|stop|>": 0.9301300418778928,
"eval_accuracy_total_num_<|content|>": 5362,
"eval_accuracy_total_num_<|from|>": 791,
"eval_accuracy_total_num_<|recipient|>": 791,
"eval_accuracy_total_num_<|stop|>": 4537,
"eval_loss": NaN,
"eval_perplexity": 1.0568973984577945,
"eval_runtime": 333.4139,
"eval_samples_per_second": 4.124,
"eval_steps_per_second": 0.516,
"step": 375
},
{
"epoch": 0.96,
"learning_rate": 3.3808211290284886e-08,
"loss": 0.3643,
"step": 376
},
{
"epoch": 0.97,
"learning_rate": 2.9155489554439364e-08,
"loss": 0.3772,
"step": 377
},
{
"epoch": 0.97,
"learning_rate": 2.4846123172992953e-08,
"loss": 0.4053,
"step": 378
},
{
"epoch": 0.97,
"learning_rate": 2.088040981046091e-08,
"loss": 0.4798,
"step": 379
},
{
"epoch": 0.97,
"learning_rate": 1.725862339392259e-08,
"loss": 0.524,
"step": 380
},
{
"epoch": 0.98,
"learning_rate": 1.3981014094099354e-08,
"loss": 0.4353,
"step": 381
},
{
"epoch": 0.98,
"learning_rate": 1.1047808308075059e-08,
"loss": 0.4712,
"step": 382
},
{
"epoch": 0.98,
"learning_rate": 8.459208643659122e-09,
"loss": 0.533,
"step": 383
},
{
"epoch": 0.98,
"learning_rate": 6.215393905388278e-09,
"loss": 0.2897,
"step": 384
},
{
"epoch": 0.99,
"learning_rate": 4.316519082179227e-09,
"loss": 0.4482,
"step": 385
},
{
"epoch": 0.99,
"learning_rate": 2.7627153366222014e-09,
"loss": 0.4109,
"step": 386
},
{
"epoch": 0.99,
"learning_rate": 1.5540899959187727e-09,
"loss": 0.3478,
"step": 387
},
{
"epoch": 0.99,
"learning_rate": 6.907265444716649e-10,
"loss": 0.4204,
"step": 388
},
{
"epoch": 1.0,
"learning_rate": 1.7268461811548176e-10,
"loss": 0.3867,
"step": 389
},
{
"epoch": 1.0,
"learning_rate": 0.0,
"loss": 0.3951,
"step": 390
}
],
"logging_steps": 1.0,
"max_steps": 390,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 100.0,
"total_flos": 805496420302848.0,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}