{
  "best_metric": 0.9165481328964233,
  "best_model_checkpoint": "work_dirs/internvl3_1b_walk_lora/checkpoint-495",
  "epoch": 0.5963122793252256,
  "eval_steps": 15,
  "global_step": 570,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 4.651162790697675e-07,
      "loss": 3.9167,
      "step": 1
    },
    {
      "epoch": 0.0,
      "learning_rate": 9.30232558139535e-07,
      "loss": 3.7216,
      "step": 2
    },
    {
      "epoch": 0.0,
      "learning_rate": 1.3953488372093025e-06,
      "loss": 3.8895,
      "step": 3
    },
    {
      "epoch": 0.0,
      "learning_rate": 1.86046511627907e-06,
      "loss": 3.9865,
      "step": 4
    },
    {
      "epoch": 0.01,
      "learning_rate": 2.3255813953488376e-06,
      "loss": 3.5411,
      "step": 5
    },
    {
      "epoch": 0.01,
      "learning_rate": 2.790697674418605e-06,
      "loss": 3.7881,
      "step": 6
    },
    {
      "epoch": 0.01,
      "learning_rate": 3.2558139534883724e-06,
      "loss": 3.5733,
      "step": 7
    },
    {
      "epoch": 0.01,
      "learning_rate": 3.72093023255814e-06,
      "loss": 3.4129,
      "step": 8
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.186046511627907e-06,
      "loss": 3.0793,
      "step": 9
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.651162790697675e-06,
      "loss": 2.6704,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 5.116279069767442e-06,
      "loss": 3.0705,
      "step": 11
    },
    {
      "epoch": 0.01,
      "learning_rate": 5.58139534883721e-06,
      "loss": 2.4496,
      "step": 12
    },
    {
      "epoch": 0.01,
      "learning_rate": 6.046511627906977e-06,
      "loss": 2.5531,
      "step": 13
    },
    {
      "epoch": 0.01,
      "learning_rate": 6.511627906976745e-06,
      "loss": 2.5058,
      "step": 14
    },
    {
      "epoch": 0.02,
      "learning_rate": 6.976744186046513e-06,
      "loss": 2.2803,
      "step": 15
    },
    {
      "epoch": 0.02,
      "eval_loss": 2.380812168121338,
      "eval_runtime": 306.4875,
      "eval_samples_per_second": 5.547,
      "eval_steps_per_second": 2.773,
      "step": 15
    },
    {
      "epoch": 0.02,
      "learning_rate": 7.44186046511628e-06,
      "loss": 2.358,
      "step": 16
    },
    {
      "epoch": 0.02,
      "learning_rate": 7.906976744186048e-06,
      "loss": 2.1133,
      "step": 17
    },
    {
      "epoch": 0.02,
      "learning_rate": 8.372093023255815e-06,
      "loss": 2.3785,
      "step": 18
    },
    {
      "epoch": 0.02,
      "learning_rate": 8.837209302325582e-06,
      "loss": 2.051,
      "step": 19
    },
    {
      "epoch": 0.02,
      "learning_rate": 9.30232558139535e-06,
      "loss": 1.8168,
      "step": 20
    },
    {
      "epoch": 0.02,
      "learning_rate": 9.767441860465117e-06,
      "loss": 2.0361,
      "step": 21
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.0232558139534884e-05,
      "loss": 2.0988,
      "step": 22
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.0697674418604651e-05,
      "loss": 1.8503,
      "step": 23
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.116279069767442e-05,
      "loss": 2.0808,
      "step": 24
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.1627906976744187e-05,
      "loss": 1.9016,
      "step": 25
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.2093023255813954e-05,
      "loss": 1.7763,
      "step": 26
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.2558139534883723e-05,
      "loss": 1.9512,
      "step": 27
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.302325581395349e-05,
      "loss": 1.8621,
      "step": 28
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.3488372093023257e-05,
      "loss": 1.5368,
      "step": 29
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.3953488372093025e-05,
      "loss": 1.9195,
      "step": 30
    },
    {
      "epoch": 0.03,
      "eval_loss": 1.6542388200759888,
      "eval_runtime": 307.0679,
      "eval_samples_per_second": 5.536,
      "eval_steps_per_second": 2.768,
      "step": 30
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.441860465116279e-05,
      "loss": 1.5574,
      "step": 31
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.488372093023256e-05,
      "loss": 1.5251,
      "step": 32
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.5348837209302328e-05,
      "loss": 1.5617,
      "step": 33
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.5813953488372095e-05,
      "loss": 1.4944,
      "step": 34
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.6279069767441862e-05,
      "loss": 1.442,
      "step": 35
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.674418604651163e-05,
      "loss": 1.542,
      "step": 36
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.7209302325581396e-05,
      "loss": 1.4577,
      "step": 37
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.7674418604651163e-05,
      "loss": 1.477,
      "step": 38
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.813953488372093e-05,
      "loss": 1.641,
      "step": 39
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.86046511627907e-05,
      "loss": 1.4285,
      "step": 40
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9069767441860468e-05,
      "loss": 1.2625,
      "step": 41
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9534883720930235e-05,
      "loss": 1.2509,
      "step": 42
    },
    {
      "epoch": 0.04,
      "learning_rate": 2e-05,
      "loss": 1.2406,
      "step": 43
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.046511627906977e-05,
      "loss": 1.3302,
      "step": 44
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.093023255813954e-05,
      "loss": 1.499,
      "step": 45
    },
    {
      "epoch": 0.05,
      "eval_loss": 1.3557870388031006,
      "eval_runtime": 304.1716,
      "eval_samples_per_second": 5.589,
      "eval_steps_per_second": 2.794,
      "step": 45
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.1395348837209303e-05,
      "loss": 1.3516,
      "step": 46
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.186046511627907e-05,
      "loss": 1.3749,
      "step": 47
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.232558139534884e-05,
      "loss": 1.507,
      "step": 48
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.2790697674418607e-05,
      "loss": 1.5157,
      "step": 49
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.3255813953488374e-05,
      "loss": 1.2619,
      "step": 50
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.3720930232558144e-05,
      "loss": 1.2551,
      "step": 51
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.4186046511627908e-05,
      "loss": 1.586,
      "step": 52
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.4651162790697675e-05,
      "loss": 1.4297,
      "step": 53
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.5116279069767445e-05,
      "loss": 1.1373,
      "step": 54
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.5581395348837212e-05,
      "loss": 1.3681,
      "step": 55
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.604651162790698e-05,
      "loss": 1.4136,
      "step": 56
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.6511627906976743e-05,
      "loss": 1.3521,
      "step": 57
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.6976744186046514e-05,
      "loss": 1.0214,
      "step": 58
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.744186046511628e-05,
      "loss": 1.3137,
      "step": 59
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.790697674418605e-05,
      "loss": 1.0428,
      "step": 60
    },
    {
      "epoch": 0.06,
      "eval_loss": 1.2505757808685303,
      "eval_runtime": 304.6984,
      "eval_samples_per_second": 5.579,
      "eval_steps_per_second": 2.79,
      "step": 60
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.8372093023255818e-05,
      "loss": 1.1287,
      "step": 61
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.883720930232558e-05,
      "loss": 1.1939,
      "step": 62
    },
    {
      "epoch": 0.07,
      "learning_rate": 2.930232558139535e-05,
      "loss": 1.4606,
      "step": 63
    },
    {
      "epoch": 0.07,
      "learning_rate": 2.976744186046512e-05,
      "loss": 1.1316,
      "step": 64
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.0232558139534886e-05,
      "loss": 1.2673,
      "step": 65
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.0697674418604656e-05,
      "loss": 1.3902,
      "step": 66
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.116279069767442e-05,
      "loss": 1.151,
      "step": 67
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.162790697674419e-05,
      "loss": 1.0023,
      "step": 68
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.2093023255813954e-05,
      "loss": 1.0705,
      "step": 69
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.2558139534883724e-05,
      "loss": 1.0287,
      "step": 70
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.3023255813953495e-05,
      "loss": 1.3004,
      "step": 71
    },
    {
      "epoch": 0.08,
      "learning_rate": 3.348837209302326e-05,
      "loss": 1.1513,
      "step": 72
    },
    {
      "epoch": 0.08,
      "learning_rate": 3.395348837209302e-05,
      "loss": 1.0704,
      "step": 73
    },
    {
      "epoch": 0.08,
      "learning_rate": 3.441860465116279e-05,
      "loss": 1.479,
      "step": 74
    },
    {
      "epoch": 0.08,
      "learning_rate": 3.488372093023256e-05,
      "loss": 1.0527,
      "step": 75
    },
    {
      "epoch": 0.08,
      "eval_loss": 1.1933207511901855,
      "eval_runtime": 306.1527,
      "eval_samples_per_second": 5.553,
      "eval_steps_per_second": 2.776,
      "step": 75
    },
    {
      "epoch": 0.08,
      "learning_rate": 3.5348837209302326e-05,
      "loss": 1.1885,
      "step": 76
    },
    {
      "epoch": 0.08,
      "learning_rate": 3.58139534883721e-05,
      "loss": 1.0932,
      "step": 77
    },
    {
      "epoch": 0.08,
      "learning_rate": 3.627906976744186e-05,
      "loss": 1.3255,
      "step": 78
    },
    {
      "epoch": 0.08,
      "learning_rate": 3.674418604651163e-05,
      "loss": 1.1993,
      "step": 79
    },
    {
      "epoch": 0.08,
      "learning_rate": 3.72093023255814e-05,
      "loss": 1.1773,
      "step": 80
    },
    {
      "epoch": 0.08,
      "learning_rate": 3.7674418604651165e-05,
      "loss": 1.16,
      "step": 81
    },
    {
      "epoch": 0.09,
      "learning_rate": 3.8139534883720935e-05,
      "loss": 1.1716,
      "step": 82
    },
    {
      "epoch": 0.09,
      "learning_rate": 3.8604651162790706e-05,
      "loss": 1.3089,
      "step": 83
    },
    {
      "epoch": 0.09,
      "learning_rate": 3.906976744186047e-05,
      "loss": 1.0025,
      "step": 84
    },
    {
      "epoch": 0.09,
      "learning_rate": 3.953488372093023e-05,
      "loss": 1.1443,
      "step": 85
    },
    {
      "epoch": 0.09,
      "learning_rate": 4e-05,
      "loss": 1.0572,
      "step": 86
    },
    {
      "epoch": 0.09,
      "learning_rate": 3.999998722024272e-05,
      "loss": 1.14,
      "step": 87
    },
    {
      "epoch": 0.09,
      "learning_rate": 3.999994888098721e-05,
      "loss": 1.1048,
      "step": 88
    },
    {
      "epoch": 0.09,
      "learning_rate": 3.9999884982282465e-05,
      "loss": 1.5029,
      "step": 89
    },
    {
      "epoch": 0.09,
      "learning_rate": 3.9999795524210154e-05,
      "loss": 1.0508,
      "step": 90
    },
    {
      "epoch": 0.09,
      "eval_loss": 1.1536332368850708,
      "eval_runtime": 305.6565,
      "eval_samples_per_second": 5.562,
      "eval_steps_per_second": 2.781,
      "step": 90
    },
    {
      "epoch": 0.1,
      "learning_rate": 3.999968050688459e-05,
      "loss": 1.0417,
      "step": 91
    },
    {
      "epoch": 0.1,
      "learning_rate": 3.9999539930452766e-05,
      "loss": 1.0267,
      "step": 92
    },
    {
      "epoch": 0.1,
      "learning_rate": 3.999937379509434e-05,
      "loss": 1.035,
      "step": 93
    },
    {
      "epoch": 0.1,
      "learning_rate": 3.999918210102163e-05,
      "loss": 1.1249,
      "step": 94
    },
    {
      "epoch": 0.1,
      "learning_rate": 3.999896484847961e-05,
      "loss": 1.2173,
      "step": 95
    },
    {
      "epoch": 0.1,
      "learning_rate": 3.999872203774593e-05,
      "loss": 0.9216,
      "step": 96
    },
    {
      "epoch": 0.1,
      "learning_rate": 3.9998453669130886e-05,
      "loss": 1.3556,
      "step": 97
    },
    {
      "epoch": 0.1,
      "learning_rate": 3.999815974297746e-05,
      "loss": 1.1525,
      "step": 98
    },
    {
      "epoch": 0.1,
      "learning_rate": 3.999784025966127e-05,
      "loss": 1.1309,
      "step": 99
    },
    {
      "epoch": 0.1,
      "learning_rate": 3.999749521959062e-05,
      "loss": 1.0473,
      "step": 100
    },
    {
      "epoch": 0.11,
      "learning_rate": 3.999712462320645e-05,
      "loss": 1.1397,
      "step": 101
    },
    {
      "epoch": 0.11,
      "learning_rate": 3.999672847098238e-05,
      "loss": 0.881,
      "step": 102
    },
    {
      "epoch": 0.11,
      "learning_rate": 3.999630676342468e-05,
      "loss": 1.2838,
      "step": 103
    },
    {
      "epoch": 0.11,
      "learning_rate": 3.9995859501072297e-05,
      "loss": 1.1335,
      "step": 104
    },
    {
      "epoch": 0.11,
      "learning_rate": 3.99953866844968e-05,
      "loss": 1.0985,
      "step": 105
    },
    {
      "epoch": 0.11,
      "eval_loss": 1.1296988725662231,
      "eval_runtime": 306.7861,
      "eval_samples_per_second": 5.541,
      "eval_steps_per_second": 2.771,
      "step": 105
    },
    {
      "epoch": 0.11,
      "learning_rate": 3.999488831430245e-05,
      "loss": 1.1491,
      "step": 106
    },
    {
      "epoch": 0.11,
      "learning_rate": 3.999436439112614e-05,
      "loss": 0.7817,
      "step": 107
    },
    {
      "epoch": 0.11,
      "learning_rate": 3.9993814915637454e-05,
      "loss": 1.1253,
      "step": 108
    },
    {
      "epoch": 0.11,
      "learning_rate": 3.999323988853858e-05,
      "loss": 0.9584,
      "step": 109
    },
    {
      "epoch": 0.12,
      "learning_rate": 3.9992639310564414e-05,
      "loss": 0.9343,
      "step": 110
    },
    {
      "epoch": 0.12,
      "learning_rate": 3.9992013182482455e-05,
      "loss": 1.0722,
      "step": 111
    },
    {
      "epoch": 0.12,
      "learning_rate": 3.99913615050929e-05,
      "loss": 0.8007,
      "step": 112
    },
    {
      "epoch": 0.12,
      "learning_rate": 3.999068427922857e-05,
      "loss": 0.9113,
      "step": 113
    },
    {
      "epoch": 0.12,
      "learning_rate": 3.9989981505754946e-05,
      "loss": 1.1036,
      "step": 114
    },
    {
      "epoch": 0.12,
      "learning_rate": 3.9989253185570145e-05,
      "loss": 1.2348,
      "step": 115
    },
    {
      "epoch": 0.12,
      "learning_rate": 3.9988499319604954e-05,
      "loss": 1.1815,
      "step": 116
    },
    {
      "epoch": 0.12,
      "learning_rate": 3.998771990882279e-05,
      "loss": 1.0974,
      "step": 117
    },
    {
      "epoch": 0.12,
      "learning_rate": 3.998691495421973e-05,
      "loss": 1.2759,
      "step": 118
    },
    {
      "epoch": 0.12,
      "learning_rate": 3.998608445682447e-05,
      "loss": 1.1176,
      "step": 119
    },
    {
      "epoch": 0.13,
      "learning_rate": 3.9985228417698375e-05,
      "loss": 1.0881,
      "step": 120
    },
    {
      "epoch": 0.13,
      "eval_loss": 1.0983185768127441,
      "eval_runtime": 305.7814,
      "eval_samples_per_second": 5.56,
      "eval_steps_per_second": 2.78,
      "step": 120
    },
    {
      "epoch": 0.13,
      "learning_rate": 3.998434683793544e-05,
      "loss": 1.1432,
      "step": 121
    },
    {
      "epoch": 0.13,
      "learning_rate": 3.998343971866231e-05,
      "loss": 1.049,
      "step": 122
    },
    {
      "epoch": 0.13,
      "learning_rate": 3.9982507061038256e-05,
      "loss": 1.2233,
      "step": 123
    },
    {
      "epoch": 0.13,
      "learning_rate": 3.998154886625519e-05,
      "loss": 0.9434,
      "step": 124
    },
    {
      "epoch": 0.13,
      "learning_rate": 3.9980565135537664e-05,
      "loss": 1.1738,
      "step": 125
    },
    {
      "epoch": 0.13,
      "learning_rate": 3.9979555870142855e-05,
      "loss": 1.1199,
      "step": 126
    },
    {
      "epoch": 0.13,
      "learning_rate": 3.997852107136058e-05,
      "loss": 1.3024,
      "step": 127
    },
    {
      "epoch": 0.13,
      "learning_rate": 3.997746074051331e-05,
      "loss": 0.9943,
      "step": 128
    },
    {
      "epoch": 0.13,
      "learning_rate": 3.9976374878956096e-05,
      "loss": 1.1173,
      "step": 129
    },
    {
      "epoch": 0.14,
      "learning_rate": 3.997526348807665e-05,
      "loss": 0.9545,
      "step": 130
    },
    {
      "epoch": 0.14,
      "learning_rate": 3.99741265692953e-05,
      "loss": 1.0374,
      "step": 131
    },
    {
      "epoch": 0.14,
      "learning_rate": 3.997296412406502e-05,
      "loss": 0.908,
      "step": 132
    },
    {
      "epoch": 0.14,
      "learning_rate": 3.997177615387136e-05,
      "loss": 1.1433,
      "step": 133
    },
    {
      "epoch": 0.14,
      "learning_rate": 3.9970562660232525e-05,
      "loss": 1.0295,
      "step": 134
    },
    {
      "epoch": 0.14,
      "learning_rate": 3.996932364469934e-05,
      "loss": 1.0962,
      "step": 135
    },
    {
      "epoch": 0.14,
      "eval_loss": 1.0783262252807617,
      "eval_runtime": 303.9624,
      "eval_samples_per_second": 5.593,
      "eval_steps_per_second": 2.796,
      "step": 135
    },
    {
      "epoch": 0.14,
      "learning_rate": 3.996805910885523e-05,
      "loss": 0.9025,
      "step": 136
    },
    {
      "epoch": 0.14,
      "learning_rate": 3.996676905431623e-05,
      "loss": 0.804,
      "step": 137
    },
    {
      "epoch": 0.14,
      "learning_rate": 3.996545348273102e-05,
      "loss": 1.1246,
      "step": 138
    },
    {
      "epoch": 0.15,
      "learning_rate": 3.9964112395780866e-05,
      "loss": 0.9042,
      "step": 139
    },
    {
      "epoch": 0.15,
      "learning_rate": 3.996274579517962e-05,
      "loss": 1.259,
      "step": 140
    },
    {
      "epoch": 0.15,
      "learning_rate": 3.996135368267379e-05,
      "loss": 0.8714,
      "step": 141
    },
    {
      "epoch": 0.15,
      "learning_rate": 3.995993606004246e-05,
      "loss": 1.0765,
      "step": 142
    },
    {
      "epoch": 0.15,
      "learning_rate": 3.9958492929097306e-05,
      "loss": 1.0438,
      "step": 143
    },
    {
      "epoch": 0.15,
      "learning_rate": 3.9957024291682615e-05,
      "loss": 1.0476,
      "step": 144
    },
    {
      "epoch": 0.15,
      "learning_rate": 3.995553014967528e-05,
      "loss": 0.9741,
      "step": 145
    },
    {
      "epoch": 0.15,
      "learning_rate": 3.995401050498477e-05,
      "loss": 1.0862,
      "step": 146
    },
    {
      "epoch": 0.15,
      "learning_rate": 3.995246535955315e-05,
      "loss": 1.1516,
      "step": 147
    },
    {
      "epoch": 0.15,
      "learning_rate": 3.99508947153551e-05,
      "loss": 0.994,
      "step": 148
    },
    {
      "epoch": 0.16,
      "learning_rate": 3.994929857439784e-05,
      "loss": 1.059,
      "step": 149
    },
    {
      "epoch": 0.16,
      "learning_rate": 3.9947676938721206e-05,
      "loss": 0.9952,
      "step": 150
    },
    {
      "epoch": 0.16,
      "eval_loss": 1.063291072845459,
      "eval_runtime": 304.2526,
      "eval_samples_per_second": 5.587,
      "eval_steps_per_second": 2.794,
      "step": 150
    },
    {
      "epoch": 0.16,
      "learning_rate": 3.994602981039761e-05,
      "loss": 1.1146,
      "step": 151
    },
    {
      "epoch": 0.16,
      "learning_rate": 3.994435719153206e-05,
      "loss": 1.0162,
      "step": 152
    },
    {
      "epoch": 0.16,
      "learning_rate": 3.9942659084262086e-05,
      "loss": 1.0621,
      "step": 153
    },
    {
      "epoch": 0.16,
      "learning_rate": 3.9940935490757864e-05,
      "loss": 0.8203,
      "step": 154
    },
    {
      "epoch": 0.16,
      "learning_rate": 3.993918641322208e-05,
      "loss": 0.9784,
      "step": 155
    },
    {
      "epoch": 0.16,
      "learning_rate": 3.993741185389003e-05,
      "loss": 1.2086,
      "step": 156
    },
    {
      "epoch": 0.16,
      "learning_rate": 3.993561181502954e-05,
      "loss": 0.9117,
      "step": 157
    },
    {
      "epoch": 0.17,
      "learning_rate": 3.9933786298941036e-05,
      "loss": 1.0186,
      "step": 158
    },
    {
      "epoch": 0.17,
      "learning_rate": 3.993193530795747e-05,
      "loss": 1.026,
      "step": 159
    },
    {
      "epoch": 0.17,
      "learning_rate": 3.9930058844444365e-05,
      "loss": 0.9734,
      "step": 160
    },
    {
      "epoch": 0.17,
      "learning_rate": 3.99281569107998e-05,
      "loss": 0.9314,
      "step": 161
    },
    {
      "epoch": 0.17,
      "learning_rate": 3.9926229509454393e-05,
      "loss": 1.0541,
      "step": 162
    },
    {
      "epoch": 0.17,
      "learning_rate": 3.9924276642871316e-05,
      "loss": 1.1284,
      "step": 163
    },
    {
      "epoch": 0.17,
      "learning_rate": 3.992229831354629e-05,
      "loss": 1.168,
      "step": 164
    },
    {
      "epoch": 0.17,
      "learning_rate": 3.9920294524007576e-05,
      "loss": 1.0302,
      "step": 165
    },
    {
      "epoch": 0.17,
      "eval_loss": 1.056298851966858,
      "eval_runtime": 303.6954,
      "eval_samples_per_second": 5.598,
      "eval_steps_per_second": 2.799,
      "step": 165
    },
    {
      "epoch": 0.17,
      "learning_rate": 3.991826527681597e-05,
      "loss": 0.9554,
      "step": 166
    },
    {
      "epoch": 0.17,
      "learning_rate": 3.9916210574564786e-05,
      "loss": 1.0582,
      "step": 167
    },
    {
      "epoch": 0.18,
      "learning_rate": 3.991413041987989e-05,
      "loss": 1.1937,
      "step": 168
    },
    {
      "epoch": 0.18,
      "learning_rate": 3.991202481541968e-05,
      "loss": 0.8932,
      "step": 169
    },
    {
      "epoch": 0.18,
      "learning_rate": 3.990989376387505e-05,
      "loss": 1.1288,
      "step": 170
    },
    {
      "epoch": 0.18,
      "learning_rate": 3.990773726796944e-05,
      "loss": 0.7828,
      "step": 171
    },
    {
      "epoch": 0.18,
      "learning_rate": 3.99055553304588e-05,
      "loss": 1.1117,
      "step": 172
    },
    {
      "epoch": 0.18,
      "learning_rate": 3.9903347954131607e-05,
      "loss": 0.8938,
      "step": 173
    },
    {
      "epoch": 0.18,
      "learning_rate": 3.9901115141808803e-05,
      "loss": 1.4246,
      "step": 174
    },
    {
      "epoch": 0.18,
      "learning_rate": 3.98988568963439e-05,
      "loss": 0.8903,
      "step": 175
    },
    {
      "epoch": 0.18,
      "learning_rate": 3.9896573220622856e-05,
      "loss": 1.1955,
      "step": 176
    },
    {
      "epoch": 0.19,
      "learning_rate": 3.9894264117564165e-05,
      "loss": 1.145,
      "step": 177
    },
    {
      "epoch": 0.19,
      "learning_rate": 3.989192959011881e-05,
      "loss": 0.932,
      "step": 178
    },
    {
      "epoch": 0.19,
      "learning_rate": 3.988956964127025e-05,
      "loss": 1.0969,
      "step": 179
    },
    {
      "epoch": 0.19,
      "learning_rate": 3.988718427403444e-05,
      "loss": 1.0822,
      "step": 180
    },
    {
      "epoch": 0.19,
      "eval_loss": 1.0386698246002197,
      "eval_runtime": 304.0194,
      "eval_samples_per_second": 5.592,
      "eval_steps_per_second": 2.796,
      "step": 180
    },
    {
      "epoch": 0.19,
      "learning_rate": 3.988477349145983e-05,
      "loss": 1.124,
      "step": 181
    },
    {
      "epoch": 0.19,
      "learning_rate": 3.988233729662735e-05,
      "loss": 1.0855,
      "step": 182
    },
    {
      "epoch": 0.19,
      "learning_rate": 3.987987569265037e-05,
      "loss": 1.1742,
      "step": 183
    },
    {
      "epoch": 0.19,
      "learning_rate": 3.9877388682674795e-05,
      "loss": 1.191,
      "step": 184
    },
    {
      "epoch": 0.19,
      "learning_rate": 3.9874876269878935e-05,
      "loss": 1.0664,
      "step": 185
    },
    {
      "epoch": 0.19,
      "learning_rate": 3.98723384574736e-05,
      "loss": 1.0231,
      "step": 186
    },
    {
      "epoch": 0.2,
      "learning_rate": 3.986977524870207e-05,
      "loss": 1.1294,
      "step": 187
    },
    {
      "epoch": 0.2,
      "learning_rate": 3.9867186646840036e-05,
      "loss": 1.0238,
      "step": 188
    },
    {
      "epoch": 0.2,
      "learning_rate": 3.986457265519568e-05,
      "loss": 1.0379,
      "step": 189
    },
    {
      "epoch": 0.2,
      "learning_rate": 3.986193327710963e-05,
      "loss": 0.7904,
      "step": 190
    },
    {
      "epoch": 0.2,
      "learning_rate": 3.985926851595493e-05,
      "loss": 1.2695,
      "step": 191
    },
    {
      "epoch": 0.2,
      "learning_rate": 3.98565783751371e-05,
      "loss": 1.0359,
      "step": 192
    },
    {
      "epoch": 0.2,
      "learning_rate": 3.9853862858094057e-05,
      "loss": 1.1448,
      "step": 193
    },
    {
      "epoch": 0.2,
      "learning_rate": 3.9851121968296176e-05,
      "loss": 1.3462,
      "step": 194
    },
    {
      "epoch": 0.2,
      "learning_rate": 3.984835570924624e-05,
      "loss": 0.9218,
      "step": 195
    },
    {
      "epoch": 0.2,
      "eval_loss": 1.0221573114395142,
      "eval_runtime": 303.9691,
      "eval_samples_per_second": 5.593,
      "eval_steps_per_second": 2.796,
      "step": 195
    },
    {
      "epoch": 0.21,
      "learning_rate": 3.9845564084479466e-05,
      "loss": 1.073,
      "step": 196
    },
    {
      "epoch": 0.21,
      "learning_rate": 3.9842747097563477e-05,
      "loss": 0.9905,
      "step": 197
    },
    {
      "epoch": 0.21,
      "learning_rate": 3.983990475209832e-05,
      "loss": 0.9532,
      "step": 198
    },
    {
      "epoch": 0.21,
      "learning_rate": 3.983703705171645e-05,
      "loss": 0.9368,
      "step": 199
    },
    {
      "epoch": 0.21,
      "learning_rate": 3.98341440000827e-05,
      "loss": 1.1203,
      "step": 200
    },
    {
      "epoch": 0.21,
      "learning_rate": 3.9831225600894335e-05,
      "loss": 1.0358,
      "step": 201
    },
    {
      "epoch": 0.21,
      "learning_rate": 3.9828281857880994e-05,
      "loss": 0.995,
      "step": 202
    },
    {
      "epoch": 0.21,
      "learning_rate": 3.982531277480471e-05,
      "loss": 0.8318,
      "step": 203
    },
    {
      "epoch": 0.21,
      "learning_rate": 3.98223183554599e-05,
      "loss": 1.0344,
      "step": 204
    },
    {
      "epoch": 0.21,
      "learning_rate": 3.981929860367335e-05,
      "loss": 1.0772,
      "step": 205
    },
    {
      "epoch": 0.22,
      "learning_rate": 3.981625352330424e-05,
      "loss": 0.9885,
      "step": 206
    },
    {
      "epoch": 0.22,
      "learning_rate": 3.9813183118244106e-05,
      "loss": 0.8655,
      "step": 207
    },
    {
      "epoch": 0.22,
      "learning_rate": 3.981008739241685e-05,
      "loss": 1.111,
      "step": 208
    },
    {
      "epoch": 0.22,
      "learning_rate": 3.980696634977873e-05,
      "loss": 0.9456,
      "step": 209
    },
    {
      "epoch": 0.22,
      "learning_rate": 3.9803819994318375e-05,
      "loss": 1.1261,
      "step": 210
    },
    {
      "epoch": 0.22,
      "eval_loss": 1.0027164220809937,
      "eval_runtime": 304.8813,
      "eval_samples_per_second": 5.576,
      "eval_steps_per_second": 2.788,
      "step": 210
    },
    {
      "epoch": 0.22,
      "learning_rate": 3.980064833005674e-05,
      "loss": 1.0561,
      "step": 211
    },
    {
      "epoch": 0.22,
      "learning_rate": 3.979745136104714e-05,
      "loss": 1.1362,
      "step": 212
    },
    {
      "epoch": 0.22,
      "learning_rate": 3.979422909137522e-05,
      "loss": 1.0523,
      "step": 213
    },
    {
      "epoch": 0.22,
      "learning_rate": 3.9790981525158964e-05,
      "loss": 1.0797,
      "step": 214
    },
    {
      "epoch": 0.22,
      "learning_rate": 3.978770866654869e-05,
      "loss": 1.1345,
      "step": 215
    },
    {
      "epoch": 0.23,
      "learning_rate": 3.9784410519727015e-05,
      "loss": 1.1091,
      "step": 216
    },
    {
      "epoch": 0.23,
      "learning_rate": 3.978108708890891e-05,
      "loss": 1.1692,
      "step": 217
    },
    {
      "epoch": 0.23,
      "learning_rate": 3.9777738378341626e-05,
      "loss": 1.1035,
      "step": 218
    },
    {
      "epoch": 0.23,
      "learning_rate": 3.977436439230474e-05,
      "loss": 0.9682,
      "step": 219
    },
    {
      "epoch": 0.23,
      "learning_rate": 3.977096513511012e-05,
      "loss": 1.0829,
      "step": 220
    },
    {
      "epoch": 0.23,
      "learning_rate": 3.976754061110193e-05,
      "loss": 1.0018,
      "step": 221
    },
    {
      "epoch": 0.23,
      "learning_rate": 3.976409082465665e-05,
      "loss": 1.0959,
      "step": 222
    },
    {
      "epoch": 0.23,
      "learning_rate": 3.9760615780183e-05,
      "loss": 0.9708,
      "step": 223
    },
    {
      "epoch": 0.23,
      "learning_rate": 3.9757115482122006e-05,
      "loss": 1.0201,
      "step": 224
    },
    {
      "epoch": 0.24,
      "learning_rate": 3.975358993494697e-05,
      "loss": 1.0763,
      "step": 225
    },
    {
      "epoch": 0.24,
      "eval_loss": 1.0071771144866943,
      "eval_runtime": 305.0297,
      "eval_samples_per_second": 5.573,
      "eval_steps_per_second": 2.787,
      "step": 225
    },
    {
      "epoch": 0.24,
      "learning_rate": 3.975003914316346e-05,
      "loss": 1.1545,
      "step": 226
    },
    {
      "epoch": 0.24,
      "learning_rate": 3.974646311130929e-05,
      "loss": 0.9185,
      "step": 227
    },
    {
      "epoch": 0.24,
      "learning_rate": 3.974286184395455e-05,
      "loss": 1.0277,
      "step": 228
    },
    {
      "epoch": 0.24,
      "learning_rate": 3.973923534570157e-05,
      "loss": 1.2275,
      "step": 229
    },
    {
      "epoch": 0.24,
      "learning_rate": 3.973558362118493e-05,
      "loss": 1.2053,
      "step": 230
    },
    {
      "epoch": 0.24,
      "learning_rate": 3.973190667507144e-05,
      "loss": 1.0772,
      "step": 231
    },
    {
      "epoch": 0.24,
      "learning_rate": 3.972820451206015e-05,
      "loss": 0.9095,
      "step": 232
    },
    {
      "epoch": 0.24,
      "learning_rate": 3.972447713688233e-05,
      "loss": 1.1663,
      "step": 233
    },
    {
      "epoch": 0.24,
      "learning_rate": 3.972072455430149e-05,
      "loss": 1.1914,
      "step": 234
    },
    {
      "epoch": 0.25,
      "learning_rate": 3.971694676911332e-05,
      "loss": 0.955,
      "step": 235
    },
    {
      "epoch": 0.25,
      "learning_rate": 3.9713143786145754e-05,
      "loss": 0.9635,
      "step": 236
    },
    {
      "epoch": 0.25,
      "learning_rate": 3.97093156102589e-05,
      "loss": 1.2736,
      "step": 237
    },
    {
      "epoch": 0.25,
      "learning_rate": 3.9705462246345076e-05,
      "loss": 0.8736,
      "step": 238
    },
    {
      "epoch": 0.25,
      "learning_rate": 3.970158369932879e-05,
      "loss": 0.9139,
      "step": 239
    },
    {
      "epoch": 0.25,
      "learning_rate": 3.969767997416673e-05,
      "loss": 0.904,
      "step": 240
    },
    {
      "epoch": 0.25,
      "eval_loss": 0.9927276372909546,
      "eval_runtime": 305.7649,
      "eval_samples_per_second": 5.56,
      "eval_steps_per_second": 2.78,
      "step": 240
    },
    {
      "epoch": 0.25,
      "learning_rate": 3.969375107584777e-05,
      "loss": 0.8896,
      "step": 241
    },
    {
      "epoch": 0.25,
      "learning_rate": 3.968979700939294e-05,
      "loss": 1.0579,
      "step": 242
    },
    {
      "epoch": 0.25,
      "learning_rate": 3.9685817779855434e-05,
      "loss": 1.0181,
      "step": 243
    },
    {
      "epoch": 0.26,
      "learning_rate": 3.968181339232062e-05,
      "loss": 0.9848,
      "step": 244
    },
    {
      "epoch": 0.26,
      "learning_rate": 3.967778385190601e-05,
      "loss": 1.1867,
      "step": 245
    },
    {
      "epoch": 0.26,
      "learning_rate": 3.967372916376125e-05,
      "loss": 0.9887,
      "step": 246
    },
    {
      "epoch": 0.26,
      "learning_rate": 3.9669649333068145e-05,
      "loss": 0.9262,
      "step": 247
    },
    {
      "epoch": 0.26,
      "learning_rate": 3.96655443650406e-05,
      "loss": 1.1448,
      "step": 248
    },
    {
      "epoch": 0.26,
      "learning_rate": 3.966141426492468e-05,
      "loss": 0.94,
      "step": 249
    },
    {
      "epoch": 0.26,
      "learning_rate": 3.9657259037998554e-05,
      "loss": 1.0778,
      "step": 250
    },
    {
      "epoch": 0.26,
      "learning_rate": 3.965307868957249e-05,
      "loss": 1.0115,
      "step": 251
    },
    {
      "epoch": 0.26,
      "learning_rate": 3.964887322498888e-05,
      "loss": 0.8901,
      "step": 252
    },
    {
      "epoch": 0.26,
      "learning_rate": 3.9644642649622207e-05,
      "loss": 1.1599,
      "step": 253
    },
    {
      "epoch": 0.27,
      "learning_rate": 3.9640386968879046e-05,
      "loss": 1.0352,
      "step": 254
    },
    {
      "epoch": 0.27,
      "learning_rate": 3.963610618819804e-05,
      "loss": 0.8569,
      "step": 255
    },
    {
      "epoch": 0.27,
      "eval_loss": 0.9998227953910828,
      "eval_runtime": 305.0651,
      "eval_samples_per_second": 5.573,
      "eval_steps_per_second": 2.786,
      "step": 255
    },
    {
      "epoch": 0.27,
      "learning_rate": 3.963180031304993e-05,
      "loss": 0.9353,
      "step": 256
    },
    {
      "epoch": 0.27,
      "learning_rate": 3.962746934893753e-05,
      "loss": 1.1608,
      "step": 257
    },
    {
      "epoch": 0.27,
      "learning_rate": 3.962311330139569e-05,
      "loss": 0.9367,
      "step": 258
    },
    {
      "epoch": 0.27,
      "learning_rate": 3.9618732175991344e-05,
      "loss": 1.0189,
      "step": 259
    },
    {
      "epoch": 0.27,
      "learning_rate": 3.9614325978323453e-05,
      "loss": 0.8885,
      "step": 260
    },
    {
      "epoch": 0.27,
      "learning_rate": 3.960989471402305e-05,
      "loss": 0.8577,
      "step": 261
    },
    {
      "epoch": 0.27,
      "learning_rate": 3.960543838875317e-05,
      "loss": 1.1157,
      "step": 262
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.960095700820888e-05,
      "loss": 0.7711,
      "step": 263
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.9596450578117296e-05,
      "loss": 0.9706,
      "step": 264
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.9591919104237516e-05,
      "loss": 0.9,
      "step": 265
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.958736259236065e-05,
      "loss": 1.0676,
      "step": 266
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.958278104830982e-05,
      "loss": 0.8822,
      "step": 267
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.957817447794011e-05,
      "loss": 0.9276,
      "step": 268
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.957354288713863e-05,
      "loss": 0.9745,
      "step": 269
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.956888628182442e-05,
      "loss": 0.8937,
      "step": 270
    },
    {
      "epoch": 0.28,
      "eval_loss": 0.9849594831466675,
      "eval_runtime": 305.1634,
      "eval_samples_per_second": 5.571,
      "eval_steps_per_second": 2.785,
      "step": 270
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.9564204667948506e-05,
      "loss": 0.8733,
      "step": 271
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.95594980514939e-05,
      "loss": 0.9085,
      "step": 272
    },
    {
      "epoch": 0.29,
      "learning_rate": 3.9554766438475526e-05,
      "loss": 1.131,
      "step": 273
    },
    {
      "epoch": 0.29,
      "learning_rate": 3.955000983494028e-05,
      "loss": 0.8823,
      "step": 274
    },
    {
      "epoch": 0.29,
      "learning_rate": 3.9545228246966976e-05,
      "loss": 0.9702,
      "step": 275
    },
    {
      "epoch": 0.29,
      "learning_rate": 3.9540421680666375e-05,
      "loss": 1.0037,
      "step": 276
    },
    {
      "epoch": 0.29,
      "learning_rate": 3.953559014218114e-05,
      "loss": 0.9346,
      "step": 277
    },
    {
      "epoch": 0.29,
      "learning_rate": 3.9530733637685876e-05,
      "loss": 0.8902,
      "step": 278
    },
    {
      "epoch": 0.29,
      "learning_rate": 3.952585217338707e-05,
      "loss": 0.8125,
      "step": 279
    },
    {
      "epoch": 0.29,
      "learning_rate": 3.952094575552312e-05,
      "loss": 0.9895,
      "step": 280
    },
    {
      "epoch": 0.29,
      "learning_rate": 3.9516014390364306e-05,
      "loss": 0.8237,
      "step": 281
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.951105808421279e-05,
      "loss": 0.877,
      "step": 282
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.9506076843402603e-05,
      "loss": 0.9399,
      "step": 283
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.950107067429967e-05,
      "loss": 0.9049,
      "step": 284
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.949603958330175e-05,
      "loss": 0.9688,
      "step": 285
    },
    {
      "epoch": 0.3,
      "eval_loss": 0.9738491177558899,
      "eval_runtime": 308.9119,
      "eval_samples_per_second": 5.503,
      "eval_steps_per_second": 2.752,
      "step": 285
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.949098357683843e-05,
      "loss": 0.6467,
      "step": 286
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.948590266137119e-05,
      "loss": 1.1168,
      "step": 287
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.948079684339331e-05,
      "loss": 0.8498,
      "step": 288
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.94756661294299e-05,
      "loss": 0.9051,
      "step": 289
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.947051052603789e-05,
      "loss": 0.8937,
      "step": 290
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.946533003980601e-05,
      "loss": 1.1148,
      "step": 291
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.9460124677354797e-05,
      "loss": 1.139,
      "step": 292
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.945489444533658e-05,
      "loss": 0.9265,
      "step": 293
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.944963935043548e-05,
      "loss": 0.9947,
      "step": 294
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.944435939936736e-05,
      "loss": 0.9887,
      "step": 295
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.943905459887989e-05,
      "loss": 0.8973,
      "step": 296
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.943372495575245e-05,
      "loss": 0.961,
      "step": 297
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.942837047679622e-05,
      "loss": 1.0102,
      "step": 298
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.942299116885409e-05,
      "loss": 0.9487,
      "step": 299
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.9417587038800666e-05,
      "loss": 1.0122,
      "step": 300
    },
    {
      "epoch": 0.31,
      "eval_loss": 0.9851600527763367,
      "eval_runtime": 310.4286,
      "eval_samples_per_second": 5.476,
      "eval_steps_per_second": 2.738,
      "step": 300
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.941215809354231e-05,
      "loss": 0.8358,
      "step": 301
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.9406704340017085e-05,
      "loss": 0.8969,
      "step": 302
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.940122578519475e-05,
      "loss": 0.7069,
      "step": 303
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.939572243607676e-05,
      "loss": 0.9926,
      "step": 304
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.9390194299696274e-05,
      "loss": 0.776,
      "step": 305
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.938464138311811e-05,
      "loss": 0.7522,
      "step": 306
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.937906369343876e-05,
      "loss": 0.8338,
      "step": 307
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.937346123778638e-05,
      "loss": 0.8795,
      "step": 308
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.936783402332076e-05,
      "loss": 0.9388,
      "step": 309
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.936218205723336e-05,
      "loss": 1.074,
      "step": 310
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.9356505346747244e-05,
      "loss": 0.9102,
      "step": 311
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.9350803899117116e-05,
      "loss": 0.742,
      "step": 312
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.934507772162928e-05,
      "loss": 0.7209,
      "step": 313
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.9339326821601666e-05,
      "loss": 0.9894,
      "step": 314
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.9333551206383764e-05,
      "loss": 1.1615,
      "step": 315
    },
    {
      "epoch": 0.33,
      "eval_loss": 0.9708228707313538,
      "eval_runtime": 307.4387,
      "eval_samples_per_second": 5.53,
      "eval_steps_per_second": 2.765,
      "step": 315
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.932775088335669e-05,
      "loss": 0.8703,
      "step": 316
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.9321925859933105e-05,
      "loss": 1.001,
      "step": 317
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.931607614355726e-05,
      "loss": 0.978,
      "step": 318
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.9310201741704933e-05,
      "loss": 0.8977,
      "step": 319
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.930430266188348e-05,
      "loss": 0.9107,
      "step": 320
    },
    {
      "epoch": 0.34,
      "learning_rate": 3.9298378911631776e-05,
      "loss": 1.0195,
      "step": 321
    },
    {
      "epoch": 0.34,
      "learning_rate": 3.929243049852023e-05,
      "loss": 0.9892,
      "step": 322
    },
    {
      "epoch": 0.34,
      "learning_rate": 3.928645743015077e-05,
      "loss": 0.9966,
      "step": 323
    },
    {
      "epoch": 0.34,
      "learning_rate": 3.9280459714156835e-05,
      "loss": 1.0334,
      "step": 324
    },
    {
      "epoch": 0.34,
      "learning_rate": 3.927443735820336e-05,
      "loss": 0.9775,
      "step": 325
    },
    {
      "epoch": 0.34,
      "learning_rate": 3.926839036998676e-05,
      "loss": 0.9655,
      "step": 326
    },
    {
      "epoch": 0.34,
      "learning_rate": 3.9262318757234954e-05,
      "loss": 0.9824,
      "step": 327
    },
    {
      "epoch": 0.34,
      "learning_rate": 3.925622252770731e-05,
      "loss": 0.8625,
      "step": 328
    },
    {
      "epoch": 0.34,
      "learning_rate": 3.9250101689194654e-05,
      "loss": 0.8367,
      "step": 329
    },
    {
      "epoch": 0.35,
      "learning_rate": 3.924395624951928e-05,
      "loss": 0.7815,
      "step": 330
    },
    {
      "epoch": 0.35,
      "eval_loss": 0.9727862477302551,
      "eval_runtime": 306.9177,
      "eval_samples_per_second": 5.539,
      "eval_steps_per_second": 2.769,
      "step": 330
    },
    {
      "epoch": 0.35,
      "learning_rate": 3.9237786216534914e-05,
      "loss": 0.9959,
      "step": 331
    },
    {
      "epoch": 0.35,
      "learning_rate": 3.923159159812669e-05,
      "loss": 1.1475,
      "step": 332
    },
    {
      "epoch": 0.35,
      "learning_rate": 3.92253724022112e-05,
      "loss": 0.9701,
      "step": 333
    },
    {
      "epoch": 0.35,
      "learning_rate": 3.9219128636736413e-05,
      "loss": 0.9098,
      "step": 334
    },
    {
      "epoch": 0.35,
      "learning_rate": 3.9212860309681706e-05,
      "loss": 0.8714,
      "step": 335
    },
    {
      "epoch": 0.35,
      "learning_rate": 3.9206567429057866e-05,
      "loss": 0.9181,
      "step": 336
    },
    {
      "epoch": 0.35,
      "learning_rate": 3.9200250002907025e-05,
      "loss": 0.8804,
      "step": 337
    },
    {
      "epoch": 0.35,
      "learning_rate": 3.919390803930271e-05,
      "loss": 0.9004,
      "step": 338
    },
    {
      "epoch": 0.35,
      "learning_rate": 3.918754154634979e-05,
      "loss": 0.7742,
      "step": 339
    },
    {
      "epoch": 0.36,
      "learning_rate": 3.91811505321845e-05,
      "loss": 1.0795,
      "step": 340
    },
    {
      "epoch": 0.36,
      "learning_rate": 3.917473500497438e-05,
      "loss": 0.7658,
      "step": 341
    },
    {
      "epoch": 0.36,
      "learning_rate": 3.916829497291834e-05,
      "loss": 0.9203,
      "step": 342
    },
    {
      "epoch": 0.36,
      "learning_rate": 3.916183044424657e-05,
      "loss": 0.8532,
      "step": 343
    },
    {
      "epoch": 0.36,
      "learning_rate": 3.9155341427220596e-05,
      "loss": 1.0076,
      "step": 344
    },
    {
      "epoch": 0.36,
      "learning_rate": 3.91488279301332e-05,
      "loss": 0.9221,
      "step": 345
    },
    {
      "epoch": 0.36,
      "eval_loss": 0.967799961566925,
      "eval_runtime": 306.3872,
      "eval_samples_per_second": 5.549,
      "eval_steps_per_second": 2.774,
      "step": 345
    },
    {
      "epoch": 0.36,
      "learning_rate": 3.9142289961308496e-05,
      "loss": 0.8362,
      "step": 346
    },
    {
      "epoch": 0.36,
      "learning_rate": 3.913572752910185e-05,
      "loss": 0.8222,
      "step": 347
    },
    {
      "epoch": 0.36,
      "learning_rate": 3.912914064189987e-05,
      "loss": 0.9317,
      "step": 348
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.912252930812045e-05,
      "loss": 0.9229,
      "step": 349
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.911589353621271e-05,
      "loss": 1.0065,
      "step": 350
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.910923333465702e-05,
      "loss": 0.7981,
      "step": 351
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.910254871196494e-05,
      "loss": 1.0924,
      "step": 352
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.909583967667927e-05,
      "loss": 0.7513,
      "step": 353
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.908910623737398e-05,
      "loss": 1.1365,
      "step": 354
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.908234840265424e-05,
      "loss": 0.9826,
      "step": 355
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.9075566181156416e-05,
      "loss": 1.0868,
      "step": 356
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.906875958154801e-05,
      "loss": 0.9694,
      "step": 357
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.906192861252769e-05,
      "loss": 1.0107,
      "step": 358
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.9055073282825274e-05,
      "loss": 0.6974,
      "step": 359
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.904819360120171e-05,
      "loss": 0.9738,
      "step": 360
    },
    {
      "epoch": 0.38,
      "eval_loss": 0.9607006907463074,
      "eval_runtime": 306.1549,
      "eval_samples_per_second": 5.553,
      "eval_steps_per_second": 2.776,
      "step": 360
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.9041289576449057e-05,
      "loss": 0.8049,
      "step": 361
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.90343612173905e-05,
      "loss": 0.8073,
      "step": 362
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.902740853288029e-05,
      "loss": 0.9288,
      "step": 363
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.9020431531803817e-05,
      "loss": 0.9355,
      "step": 364
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.9013430223077514e-05,
      "loss": 0.8972,
      "step": 365
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.9006404615648866e-05,
      "loss": 0.9281,
      "step": 366
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.899935471849645e-05,
      "loss": 1.2577,
      "step": 367
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.899228054062985e-05,
      "loss": 0.841,
      "step": 368
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.89851820910897e-05,
      "loss": 0.8895,
      "step": 369
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.897805937894764e-05,
      "loss": 0.8683,
      "step": 370
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.897091241330633e-05,
      "loss": 0.9256,
      "step": 371
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.896374120329942e-05,
      "loss": 0.9269,
      "step": 372
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.895654575809153e-05,
      "loss": 0.8986,
      "step": 373
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.894932608687827e-05,
      "loss": 0.9923,
      "step": 374
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.8942082198886214e-05,
      "loss": 0.8926,
      "step": 375
    },
    {
      "epoch": 0.39,
      "eval_loss": 0.9774990081787109,
      "eval_runtime": 305.9785,
      "eval_samples_per_second": 5.556,
      "eval_steps_per_second": 2.778,
      "step": 375
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.8934814103372874e-05,
      "loss": 0.9404,
      "step": 376
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.892752180962668e-05,
      "loss": 0.8682,
      "step": 377
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.8920205326967024e-05,
      "loss": 0.9369,
      "step": 378
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.89128646647442e-05,
      "loss": 0.9146,
      "step": 379
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.890549983233938e-05,
      "loss": 0.8491,
      "step": 380
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.889811083916464e-05,
      "loss": 0.6624,
      "step": 381
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.8890697694662944e-05,
      "loss": 0.8896,
      "step": 382
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.888326040830811e-05,
      "loss": 0.835,
      "step": 383
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.88757989896048e-05,
      "loss": 1.0182,
      "step": 384
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.8868313448088544e-05,
      "loss": 1.0434,
      "step": 385
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.8860803793325665e-05,
      "loss": 1.0036,
      "step": 386
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.885327003491332e-05,
      "loss": 0.8763,
      "step": 387
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.8845712182479485e-05,
      "loss": 1.0394,
      "step": 388
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.88381302456829e-05,
      "loss": 1.0175,
      "step": 389
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.883052423421309e-05,
      "loss": 0.8687,
      "step": 390
    },
    {
      "epoch": 0.41,
      "eval_loss": 0.9447183609008789,
      "eval_runtime": 306.605,
      "eval_samples_per_second": 5.545,
      "eval_steps_per_second": 2.772,
      "step": 390
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.8822894157790374e-05,
      "loss": 1.248,
      "step": 391
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.881524002616578e-05,
      "loss": 0.9942,
      "step": 392
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.8807561849121125e-05,
      "loss": 0.8235,
      "step": 393
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.8799859636468914e-05,
      "loss": 0.9591,
      "step": 394
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.8792133398052394e-05,
      "loss": 1.0208,
      "step": 395
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.878438314374552e-05,
      "loss": 1.0628,
      "step": 396
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.877660888345292e-05,
      "loss": 0.9475,
      "step": 397
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.8768810627109904e-05,
      "loss": 0.8035,
      "step": 398
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.8760988384682464e-05,
      "loss": 0.9432,
      "step": 399
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.875314216616723e-05,
      "loss": 0.8064,
      "step": 400
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.874527198159149e-05,
      "loss": 0.8382,
      "step": 401
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.8737377841013135e-05,
      "loss": 0.8726,
      "step": 402
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.872945975452069e-05,
      "loss": 0.8602,
      "step": 403
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.872151773223328e-05,
      "loss": 0.8524,
      "step": 404
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.871355178430062e-05,
      "loss": 0.8231,
      "step": 405
    },
    {
      "epoch": 0.42,
      "eval_loss": 0.9651253819465637,
      "eval_runtime": 305.6205,
      "eval_samples_per_second": 5.562,
      "eval_steps_per_second": 2.781,
      "step": 405
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.870556192090299e-05,
      "loss": 1.1746,
      "step": 406
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.869754815225124e-05,
      "loss": 0.8782,
      "step": 407
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.868951048858678e-05,
      "loss": 0.7973,
      "step": 408
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.868144894018155e-05,
      "loss": 0.9944,
      "step": 409
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.8673363517338004e-05,
      "loss": 0.8298,
      "step": 410
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.866525423038912e-05,
      "loss": 0.9659,
      "step": 411
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.8657121089698374e-05,
      "loss": 0.9362,
      "step": 412
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.8648964105659713e-05,
      "loss": 0.844,
      "step": 413
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.864078328869758e-05,
      "loss": 0.9089,
      "step": 414
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.863257864926684e-05,
      "loss": 0.8935,
      "step": 415
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.862435019785284e-05,
      "loss": 1.1008,
      "step": 416
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.861609794497134e-05,
      "loss": 1.0733,
      "step": 417
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.860782190116851e-05,
      "loss": 0.8575,
      "step": 418
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.859952207702093e-05,
      "loss": 0.9333,
      "step": 419
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.859119848313559e-05,
      "loss": 1.0578,
      "step": 420
    },
    {
      "epoch": 0.44,
      "eval_loss": 0.9548206329345703,
      "eval_runtime": 305.7907,
      "eval_samples_per_second": 5.559,
      "eval_steps_per_second": 2.78,
      "step": 420
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.858285113014983e-05,
      "loss": 0.9433,
      "step": 421
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.8574480028731364e-05,
      "loss": 0.7995,
      "step": 422
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.856608518957827e-05,
      "loss": 1.1884,
      "step": 423
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.8557666623418925e-05,
      "loss": 0.8271,
      "step": 424
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.854922434101207e-05,
      "loss": 0.6884,
      "step": 425
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.854075835314674e-05,
      "loss": 0.9837,
      "step": 426
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.853226867064224e-05,
      "loss": 0.7566,
      "step": 427
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.85237553043482e-05,
      "loss": 1.0066,
      "step": 428
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.851521826514448e-05,
      "loss": 0.931,
      "step": 429
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.850665756394123e-05,
      "loss": 0.6008,
      "step": 430
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.8498073211678795e-05,
      "loss": 0.885,
      "step": 431
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.8489465219327786e-05,
      "loss": 0.9345,
      "step": 432
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.8480833597888997e-05,
      "loss": 0.794,
      "step": 433
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.847217835839344e-05,
      "loss": 0.6602,
      "step": 434
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.846349951190228e-05,
      "loss": 0.7165,
      "step": 435
    },
    {
      "epoch": 0.46,
      "eval_loss": 0.9528548121452332,
      "eval_runtime": 305.484,
      "eval_samples_per_second": 5.565,
      "eval_steps_per_second": 2.782,
      "step": 435
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.84547970695069e-05,
      "loss": 1.0008,
      "step": 436
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.8446071042328796e-05,
      "loss": 0.7078,
      "step": 437
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.8437321441519627e-05,
      "loss": 0.7031,
      "step": 438
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.8428548278261156e-05,
      "loss": 0.8728,
      "step": 439
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.841975156376529e-05,
      "loss": 0.8652,
      "step": 440
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.8410931309273996e-05,
      "loss": 1.0756,
      "step": 441
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.840208752605937e-05,
      "loss": 1.0142,
      "step": 442
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.8393220225423526e-05,
      "loss": 0.9414,
      "step": 443
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.8384329418698676e-05,
      "loss": 0.9225,
      "step": 444
    },
    {
      "epoch": 0.47,
      "learning_rate": 3.837541511724705e-05,
      "loss": 0.8286,
      "step": 445
    },
    {
      "epoch": 0.47,
      "learning_rate": 3.8366477332460916e-05,
      "loss": 0.7936,
      "step": 446
    },
    {
      "epoch": 0.47,
      "learning_rate": 3.835751607576254e-05,
      "loss": 0.7526,
      "step": 447
    },
    {
      "epoch": 0.47,
      "learning_rate": 3.834853135860419e-05,
      "loss": 1.0087,
      "step": 448
    },
    {
      "epoch": 0.47,
      "learning_rate": 3.833952319246812e-05,
      "loss": 0.8323,
      "step": 449
    },
    {
      "epoch": 0.47,
      "learning_rate": 3.8330491588866545e-05,
      "loss": 0.8661,
      "step": 450
    },
    {
      "epoch": 0.47,
      "eval_loss": 0.9453563094139099,
      "eval_runtime": 305.271,
      "eval_samples_per_second": 5.569,
      "eval_steps_per_second": 2.784,
      "step": 450
    },
    {
      "epoch": 0.47,
      "learning_rate": 3.832143655934163e-05,
      "loss": 0.9986,
      "step": 451
    },
    {
      "epoch": 0.47,
      "learning_rate": 3.8312358115465486e-05,
      "loss": 1.0959,
      "step": 452
    },
    {
      "epoch": 0.47,
      "learning_rate": 3.8303256268840156e-05,
      "loss": 0.8215,
      "step": 453
    },
    {
      "epoch": 0.47,
      "learning_rate": 3.829413103109756e-05,
      "loss": 0.8707,
      "step": 454
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.828498241389954e-05,
      "loss": 1.0806,
      "step": 455
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.8275810428937814e-05,
      "loss": 0.9381,
      "step": 456
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.826661508793395e-05,
      "loss": 0.9628,
      "step": 457
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.825739640263936e-05,
      "loss": 0.8614,
      "step": 458
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.824815438483531e-05,
      "loss": 1.013,
      "step": 459
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.823888904633288e-05,
      "loss": 0.7337,
      "step": 460
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.822960039897293e-05,
      "loss": 1.0546,
      "step": 461
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.822028845462615e-05,
      "loss": 1.0693,
      "step": 462
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.8210953225192954e-05,
      "loss": 0.9549,
      "step": 463
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.820159472260356e-05,
      "loss": 0.8109,
      "step": 464
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.8192212958817894e-05,
      "loss": 0.8778,
      "step": 465
    },
    {
      "epoch": 0.49,
      "eval_loss": 0.9310696721076965,
      "eval_runtime": 304.721,
      "eval_samples_per_second": 5.579,
      "eval_steps_per_second": 2.789,
      "step": 465
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.818280794582563e-05,
      "loss": 0.7249,
      "step": 466
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.8173379695646135e-05,
      "loss": 0.9543,
      "step": 467
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.8163928220328495e-05,
      "loss": 0.7852,
      "step": 468
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.815445353195146e-05,
      "loss": 0.8454,
      "step": 469
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.814495564262346e-05,
      "loss": 0.9077,
      "step": 470
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.813543456448256e-05,
      "loss": 0.7896,
      "step": 471
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.812589030969647e-05,
      "loss": 0.781,
      "step": 472
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.811632289046251e-05,
      "loss": 0.7486,
      "step": 473
    },
    {
      "epoch": 0.5,
      "learning_rate": 3.8106732319007604e-05,
      "loss": 0.7739,
      "step": 474
    },
    {
      "epoch": 0.5,
      "learning_rate": 3.8097118607588294e-05,
      "loss": 0.8309,
      "step": 475
    },
    {
      "epoch": 0.5,
      "learning_rate": 3.808748176849065e-05,
      "loss": 0.9477,
      "step": 476
    },
    {
      "epoch": 0.5,
      "learning_rate": 3.807782181403032e-05,
      "loss": 0.7947,
      "step": 477
    },
    {
      "epoch": 0.5,
      "learning_rate": 3.80681387565525e-05,
      "loss": 0.9982,
      "step": 478
    },
    {
      "epoch": 0.5,
      "learning_rate": 3.805843260843189e-05,
      "loss": 0.9384,
      "step": 479
    },
    {
      "epoch": 0.5,
      "learning_rate": 3.804870338207273e-05,
      "loss": 0.9454,
      "step": 480
    },
    {
      "epoch": 0.5,
      "eval_loss": 0.9295728802680969,
      "eval_runtime": 303.2716,
      "eval_samples_per_second": 5.606,
      "eval_steps_per_second": 2.803,
      "step": 480
    },
    {
      "epoch": 0.5,
      "learning_rate": 3.803895108990872e-05,
      "loss": 0.8665,
      "step": 481
    },
    {
      "epoch": 0.5,
      "learning_rate": 3.802917574440305e-05,
      "loss": 0.6593,
      "step": 482
    },
    {
      "epoch": 0.51,
      "learning_rate": 3.801937735804838e-05,
      "loss": 1.0062,
      "step": 483
    },
    {
      "epoch": 0.51,
      "learning_rate": 3.8009555943366824e-05,
      "loss": 0.7783,
      "step": 484
    },
    {
      "epoch": 0.51,
      "learning_rate": 3.799971151290989e-05,
      "loss": 0.9229,
      "step": 485
    },
    {
      "epoch": 0.51,
      "learning_rate": 3.798984407925853e-05,
      "loss": 0.8356,
      "step": 486
    },
    {
      "epoch": 0.51,
      "learning_rate": 3.7979953655023084e-05,
      "loss": 0.8971,
      "step": 487
    },
    {
      "epoch": 0.51,
      "learning_rate": 3.797004025284327e-05,
      "loss": 0.9609,
      "step": 488
    },
    {
      "epoch": 0.51,
      "learning_rate": 3.796010388538819e-05,
      "loss": 0.8788,
      "step": 489
    },
    {
      "epoch": 0.51,
      "learning_rate": 3.795014456535627e-05,
      "loss": 0.9698,
      "step": 490
    },
    {
      "epoch": 0.51,
      "learning_rate": 3.794016230547527e-05,
      "loss": 0.7551,
      "step": 491
    },
    {
      "epoch": 0.51,
      "learning_rate": 3.793015711850229e-05,
      "loss": 1.1267,
      "step": 492
    },
    {
      "epoch": 0.52,
      "learning_rate": 3.792012901722371e-05,
      "loss": 0.9498,
      "step": 493
    },
    {
      "epoch": 0.52,
      "learning_rate": 3.791007801445521e-05,
      "loss": 0.9095,
      "step": 494
    },
    {
      "epoch": 0.52,
      "learning_rate": 3.7900004123041704e-05,
      "loss": 0.9123,
      "step": 495
    },
    {
      "epoch": 0.52,
      "eval_loss": 0.9165481328964233,
      "eval_runtime": 304.8466,
      "eval_samples_per_second": 5.577,
      "eval_steps_per_second": 2.788,
      "step": 495
    },
    {
      "epoch": 0.52,
      "learning_rate": 3.788990735585741e-05,
      "loss": 0.8464,
      "step": 496
    },
    {
      "epoch": 0.52,
      "learning_rate": 3.787978772580573e-05,
      "loss": 0.9603,
      "step": 497
    },
    {
      "epoch": 0.52,
      "learning_rate": 3.786964524581931e-05,
      "loss": 0.9754,
      "step": 498
    },
    {
      "epoch": 0.52,
      "learning_rate": 3.7859479928860004e-05,
      "loss": 0.8308,
      "step": 499
    },
    {
      "epoch": 0.52,
      "learning_rate": 3.784929178791883e-05,
      "loss": 0.9541,
      "step": 500
    },
    {
      "epoch": 0.52,
      "learning_rate": 3.783908083601598e-05,
      "loss": 0.9817,
      "step": 501
    },
    {
      "epoch": 0.53,
      "learning_rate": 3.7828847086200816e-05,
      "loss": 0.7953,
      "step": 502
    },
    {
      "epoch": 0.53,
      "learning_rate": 3.781859055155182e-05,
      "loss": 1.0135,
      "step": 503
    },
    {
      "epoch": 0.53,
      "learning_rate": 3.7808311245176585e-05,
      "loss": 0.9914,
      "step": 504
    },
    {
      "epoch": 0.53,
      "learning_rate": 3.779800918021182e-05,
      "loss": 0.8723,
      "step": 505
    },
    {
      "epoch": 0.53,
      "learning_rate": 3.778768436982331e-05,
      "loss": 0.9673,
      "step": 506
    },
    {
      "epoch": 0.53,
      "learning_rate": 3.7777336827205925e-05,
      "loss": 0.9691,
      "step": 507
    },
    {
      "epoch": 0.53,
      "learning_rate": 3.7766966565583557e-05,
      "loss": 0.9857,
      "step": 508
    },
    {
      "epoch": 0.53,
      "learning_rate": 3.775657359820916e-05,
      "loss": 0.7649,
      "step": 509
    },
    {
      "epoch": 0.53,
      "learning_rate": 3.774615793836469e-05,
      "loss": 0.9766,
      "step": 510
    },
    {
      "epoch": 0.53,
      "eval_loss": 0.9254794716835022,
      "eval_runtime": 304.5773,
      "eval_samples_per_second": 5.582,
      "eval_steps_per_second": 2.791,
      "step": 510
    },
    {
      "epoch": 0.53,
      "learning_rate": 3.773571959936111e-05,
      "loss": 0.97,
      "step": 511
    },
    {
      "epoch": 0.54,
      "learning_rate": 3.7725258594538356e-05,
      "loss": 0.875,
      "step": 512
    },
    {
      "epoch": 0.54,
      "learning_rate": 3.771477493726535e-05,
      "loss": 0.8327,
      "step": 513
    },
    {
      "epoch": 0.54,
      "learning_rate": 3.770426864093995e-05,
      "loss": 0.8414,
      "step": 514
    },
    {
      "epoch": 0.54,
      "learning_rate": 3.769373971898893e-05,
      "loss": 0.8273,
      "step": 515
    },
    {
      "epoch": 0.54,
      "learning_rate": 3.768318818486801e-05,
      "loss": 0.8993,
      "step": 516
    },
    {
      "epoch": 0.54,
      "learning_rate": 3.7672614052061803e-05,
      "loss": 1.0197,
      "step": 517
    },
    {
      "epoch": 0.54,
      "learning_rate": 3.766201733408378e-05,
      "loss": 0.8864,
      "step": 518
    },
    {
      "epoch": 0.54,
      "learning_rate": 3.7651398044476295e-05,
      "loss": 0.97,
      "step": 519
    },
    {
      "epoch": 0.54,
      "learning_rate": 3.7640756196810545e-05,
      "loss": 0.8077,
      "step": 520
    },
    {
      "epoch": 0.55,
      "learning_rate": 3.763009180468655e-05,
      "loss": 0.9085,
      "step": 521
    },
    {
      "epoch": 0.55,
      "learning_rate": 3.761940488173315e-05,
      "loss": 0.8263,
      "step": 522
    },
    {
      "epoch": 0.55,
      "learning_rate": 3.7608695441607957e-05,
      "loss": 1.1356,
      "step": 523
    },
    {
      "epoch": 0.55,
      "learning_rate": 3.759796349799738e-05,
      "loss": 1.0908,
      "step": 524
    },
    {
      "epoch": 0.55,
      "learning_rate": 3.75872090646166e-05,
      "loss": 0.7691,
      "step": 525
    },
    {
      "epoch": 0.55,
      "eval_loss": 0.9344983696937561,
      "eval_runtime": 304.6471,
      "eval_samples_per_second": 5.58,
      "eval_steps_per_second": 2.79,
      "step": 525
    },
    {
      "epoch": 0.55,
      "learning_rate": 3.757643215520951e-05,
      "loss": 0.8672,
      "step": 526
    },
    {
      "epoch": 0.55,
      "learning_rate": 3.756563278354874e-05,
      "loss": 0.8334,
      "step": 527
    },
    {
      "epoch": 0.55,
      "learning_rate": 3.7554810963435626e-05,
      "loss": 0.948,
      "step": 528
    },
    {
      "epoch": 0.55,
      "learning_rate": 3.754396670870018e-05,
      "loss": 1.0321,
      "step": 529
    },
    {
      "epoch": 0.55,
      "learning_rate": 3.753310003320111e-05,
      "loss": 0.8359,
      "step": 530
    },
    {
      "epoch": 0.56,
      "learning_rate": 3.752221095082576e-05,
      "loss": 0.9983,
      "step": 531
    },
    {
      "epoch": 0.56,
      "learning_rate": 3.7511299475490115e-05,
      "loss": 0.8565,
      "step": 532
    },
    {
      "epoch": 0.56,
      "learning_rate": 3.750036562113877e-05,
      "loss": 0.7781,
      "step": 533
    },
    {
      "epoch": 0.56,
      "learning_rate": 3.748940940174493e-05,
      "loss": 0.7394,
      "step": 534
    },
    {
      "epoch": 0.56,
      "learning_rate": 3.747843083131037e-05,
      "loss": 0.7705,
      "step": 535
    },
    {
      "epoch": 0.56,
      "learning_rate": 3.746742992386544e-05,
      "loss": 0.7674,
      "step": 536
    },
    {
      "epoch": 0.56,
      "learning_rate": 3.7456406693469044e-05,
      "loss": 0.8039,
      "step": 537
    },
    {
      "epoch": 0.56,
      "learning_rate": 3.744536115420859e-05,
      "loss": 0.9742,
      "step": 538
    },
    {
      "epoch": 0.56,
      "learning_rate": 3.743429332020001e-05,
      "loss": 0.8173,
      "step": 539
    },
    {
      "epoch": 0.56,
      "learning_rate": 3.7423203205587737e-05,
      "loss": 1.0856,
      "step": 540
    },
    {
      "epoch": 0.56,
      "eval_loss": 0.9283236265182495,
      "eval_runtime": 304.7287,
      "eval_samples_per_second": 5.579,
      "eval_steps_per_second": 2.789,
      "step": 540
    },
    {
      "epoch": 0.57,
      "learning_rate": 3.741209082454466e-05,
      "loss": 0.9147,
      "step": 541
    },
    {
      "epoch": 0.57,
      "learning_rate": 3.740095619127213e-05,
      "loss": 0.9475,
      "step": 542
    },
    {
      "epoch": 0.57,
      "learning_rate": 3.738979931999994e-05,
      "loss": 0.9183,
      "step": 543
    },
    {
      "epoch": 0.57,
      "learning_rate": 3.737862022498631e-05,
      "loss": 0.8878,
      "step": 544
    },
    {
      "epoch": 0.57,
      "learning_rate": 3.736741892051784e-05,
      "loss": 0.8737,
      "step": 545
    },
    {
      "epoch": 0.57,
      "learning_rate": 3.735619542090953e-05,
      "loss": 0.9091,
      "step": 546
    },
    {
      "epoch": 0.57,
      "learning_rate": 3.734494974050474e-05,
      "loss": 0.8489,
      "step": 547
    },
    {
      "epoch": 0.57,
      "learning_rate": 3.733368189367518e-05,
      "loss": 0.9828,
      "step": 548
    },
    {
      "epoch": 0.57,
      "learning_rate": 3.732239189482088e-05,
      "loss": 0.7421,
      "step": 549
    },
    {
      "epoch": 0.58,
      "learning_rate": 3.7311079758370185e-05,
      "loss": 1.0874,
      "step": 550
    },
    {
      "epoch": 0.58,
      "learning_rate": 3.729974549877973e-05,
      "loss": 0.9415,
      "step": 551
    },
    {
      "epoch": 0.58,
      "learning_rate": 3.728838913053442e-05,
      "loss": 0.8442,
      "step": 552
    },
    {
      "epoch": 0.58,
      "learning_rate": 3.727701066814743e-05,
      "loss": 0.9817,
      "step": 553
    },
    {
      "epoch": 0.58,
      "learning_rate": 3.7265610126160155e-05,
      "loss": 1.0621,
      "step": 554
    },
    {
      "epoch": 0.58,
      "learning_rate": 3.72541875191422e-05,
      "loss": 0.8558,
      "step": 555
    },
    {
      "epoch": 0.58,
      "eval_loss": 0.9172534346580505,
      "eval_runtime": 303.7298,
      "eval_samples_per_second": 5.597,
      "eval_steps_per_second": 2.799,
      "step": 555
    },
    {
      "epoch": 0.58,
      "learning_rate": 3.724274286169139e-05,
      "loss": 0.7757,
      "step": 556
    },
    {
      "epoch": 0.58,
      "learning_rate": 3.723127616843372e-05,
      "loss": 0.7697,
      "step": 557
    },
    {
      "epoch": 0.58,
      "learning_rate": 3.721978745402334e-05,
      "loss": 0.7373,
      "step": 558
    },
    {
      "epoch": 0.58,
      "learning_rate": 3.720827673314255e-05,
      "loss": 0.8195,
      "step": 559
    },
    {
      "epoch": 0.59,
      "learning_rate": 3.719674402050177e-05,
      "loss": 0.8909,
      "step": 560
    },
    {
      "epoch": 0.59,
      "learning_rate": 3.718518933083954e-05,
      "loss": 0.7498,
      "step": 561
    },
    {
      "epoch": 0.59,
      "learning_rate": 3.717361267892245e-05,
      "loss": 0.6805,
      "step": 562
    },
    {
      "epoch": 0.59,
      "learning_rate": 3.71620140795452e-05,
      "loss": 0.7792,
      "step": 563
    },
    {
      "epoch": 0.59,
      "learning_rate": 3.715039354753051e-05,
      "loss": 0.7878,
      "step": 564
    },
    {
      "epoch": 0.59,
      "learning_rate": 3.7138751097729134e-05,
      "loss": 0.9038,
      "step": 565
    },
    {
      "epoch": 0.59,
      "learning_rate": 3.712708674501985e-05,
      "loss": 0.7687,
      "step": 566
    },
    {
      "epoch": 0.59,
      "learning_rate": 3.711540050430941e-05,
      "loss": 0.8616,
      "step": 567
    },
    {
      "epoch": 0.59,
      "learning_rate": 3.710369239053255e-05,
      "loss": 1.0138,
      "step": 568
    },
    {
      "epoch": 0.6,
      "learning_rate": 3.709196241865195e-05,
      "loss": 0.9101,
      "step": 569
    },
    {
      "epoch": 0.6,
      "learning_rate": 3.708021060365824e-05,
      "loss": 0.846,
      "step": 570
    },
    {
      "epoch": 0.6,
      "eval_loss": 0.9239464402198792,
      "eval_runtime": 303.1282,
      "eval_samples_per_second": 5.608,
      "eval_steps_per_second": 2.804,
      "step": 570
    },
    {
      "epoch": 0.6,
      "step": 570,
      "total_flos": 8.625793033436135e+19,
      "train_loss": 1.079708440157405,
      "train_runtime": 14191.3031,
      "train_samples_per_second": 3.233,
      "train_steps_per_second": 0.202
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 2865,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 15,
  "total_flos": 8.625793033436135e+19,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}