{
  "best_metric": 0.5938899517059326,
  "best_model_checkpoint": "ckpt/origin_vehicle_view/vehicle_rewrite/checkpoint-96",
  "epoch": 14.328358208955224,
  "eval_steps": 6,
  "global_step": 120,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12,
      "learning_rate": 5e-05,
      "loss": 1.4214,
      "step": 1
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001,
      "loss": 1.3775,
      "step": 2
    },
    {
      "epoch": 0.36,
      "learning_rate": 9.998228055617263e-05,
      "loss": 1.4409,
      "step": 3
    },
    {
      "epoch": 0.48,
      "learning_rate": 9.99291347838381e-05,
      "loss": 1.3779,
      "step": 4
    },
    {
      "epoch": 0.6,
      "learning_rate": 9.98406003515375e-05,
      "loss": 1.3939,
      "step": 5
    },
    {
      "epoch": 0.72,
      "learning_rate": 9.971674001050686e-05,
      "loss": 1.3003,
      "step": 6
    },
    {
      "epoch": 0.72,
      "eval_loss": 1.2484569549560547,
      "eval_runtime": 126.7576,
      "eval_samples_per_second": 1.459,
      "eval_steps_per_second": 1.459,
      "step": 6
    },
    {
      "epoch": 0.84,
      "learning_rate": 9.955764155020037e-05,
      "loss": 1.2047,
      "step": 7
    },
    {
      "epoch": 0.96,
      "learning_rate": 9.936341773606723e-05,
      "loss": 1.2169,
      "step": 8
    },
    {
      "epoch": 1.07,
      "learning_rate": 9.913420622962606e-05,
      "loss": 1.1941,
      "step": 9
    },
    {
      "epoch": 1.19,
      "learning_rate": 9.887016949089333e-05,
      "loss": 1.1275,
      "step": 10
    },
    {
      "epoch": 1.31,
      "learning_rate": 9.857149466323549e-05,
      "loss": 1.1525,
      "step": 11
    },
    {
      "epoch": 1.43,
      "learning_rate": 9.82383934407258e-05,
      "loss": 1.0776,
      "step": 12
    },
    {
      "epoch": 1.43,
      "eval_loss": 1.0902503728866577,
      "eval_runtime": 126.1532,
      "eval_samples_per_second": 1.466,
      "eval_steps_per_second": 1.466,
      "step": 12
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.787110191810027e-05,
      "loss": 1.0459,
      "step": 13
    },
    {
      "epoch": 1.67,
      "learning_rate": 9.746988042341906e-05,
      "loss": 1.0572,
      "step": 14
    },
    {
      "epoch": 1.79,
      "learning_rate": 9.703501333355168e-05,
      "loss": 0.9835,
      "step": 15
    },
    {
      "epoch": 1.91,
      "learning_rate": 9.656680887261693e-05,
      "loss": 1.0297,
      "step": 16
    },
    {
      "epoch": 2.03,
      "learning_rate": 9.606559889352064e-05,
      "loss": 0.9979,
      "step": 17
    },
    {
      "epoch": 2.15,
      "learning_rate": 9.553173864274567e-05,
      "loss": 0.9683,
      "step": 18
    },
    {
      "epoch": 2.15,
      "eval_loss": 0.9658757448196411,
      "eval_runtime": 125.5373,
      "eval_samples_per_second": 1.474,
      "eval_steps_per_second": 1.474,
      "step": 18
    },
    {
      "epoch": 2.27,
      "learning_rate": 9.496560650856097e-05,
      "loss": 0.8934,
      "step": 19
    },
    {
      "epoch": 2.39,
      "learning_rate": 9.436760375282859e-05,
      "loss": 0.9647,
      "step": 20
    },
    {
      "epoch": 2.51,
      "learning_rate": 9.373815422659806e-05,
      "loss": 0.9392,
      "step": 21
    },
    {
      "epoch": 2.63,
      "learning_rate": 9.30777040696903e-05,
      "loss": 0.8951,
      "step": 22
    },
    {
      "epoch": 2.75,
      "learning_rate": 9.238672139448354e-05,
      "loss": 0.8752,
      "step": 23
    },
    {
      "epoch": 2.87,
      "learning_rate": 9.166569595412575e-05,
      "loss": 0.8834,
      "step": 24
    },
    {
      "epoch": 2.87,
      "eval_loss": 0.855851948261261,
      "eval_runtime": 125.6401,
      "eval_samples_per_second": 1.472,
      "eval_steps_per_second": 1.472,
      "step": 24
    },
    {
      "epoch": 2.99,
      "learning_rate": 9.091513879540845e-05,
      "loss": 0.7849,
      "step": 25
    },
    {
      "epoch": 3.1,
      "learning_rate": 9.013558189654819e-05,
      "loss": 0.8131,
      "step": 26
    },
    {
      "epoch": 3.22,
      "learning_rate": 8.932757779013214e-05,
      "loss": 0.8302,
      "step": 27
    },
    {
      "epoch": 3.34,
      "learning_rate": 8.849169917149531e-05,
      "loss": 0.8371,
      "step": 28
    },
    {
      "epoch": 3.46,
      "learning_rate": 8.762853849280693e-05,
      "loss": 0.8085,
      "step": 29
    },
    {
      "epoch": 3.58,
      "learning_rate": 8.673870754315336e-05,
      "loss": 0.7759,
      "step": 30
    },
    {
      "epoch": 3.58,
      "eval_loss": 0.7806031703948975,
      "eval_runtime": 125.7778,
      "eval_samples_per_second": 1.471,
      "eval_steps_per_second": 1.471,
      "step": 30
    },
    {
      "epoch": 3.7,
      "learning_rate": 8.582283701491576e-05,
      "loss": 0.7816,
      "step": 31
    },
    {
      "epoch": 3.82,
      "learning_rate": 8.488157605674925e-05,
      "loss": 0.7651,
      "step": 32
    },
    {
      "epoch": 3.94,
      "learning_rate": 8.391559181348082e-05,
      "loss": 0.6714,
      "step": 33
    },
    {
      "epoch": 4.06,
      "learning_rate": 8.292556895325194e-05,
      "loss": 0.7076,
      "step": 34
    },
    {
      "epoch": 4.18,
      "learning_rate": 8.191220918224101e-05,
      "loss": 0.6913,
      "step": 35
    },
    {
      "epoch": 4.3,
      "learning_rate": 8.08762307473096e-05,
      "loss": 0.7119,
      "step": 36
    },
    {
      "epoch": 4.3,
      "eval_loss": 0.72127765417099,
      "eval_runtime": 125.4909,
      "eval_samples_per_second": 1.474,
      "eval_steps_per_second": 1.474,
      "step": 36
    },
    {
      "epoch": 4.42,
      "learning_rate": 7.981836792692508e-05,
      "loss": 0.6698,
      "step": 37
    },
    {
      "epoch": 4.54,
      "learning_rate": 7.873937051072035e-05,
      "loss": 0.6715,
      "step": 38
    },
    {
      "epoch": 4.66,
      "learning_rate": 7.764000326805967e-05,
      "loss": 0.6827,
      "step": 39
    },
    {
      "epoch": 4.78,
      "learning_rate": 7.652104540598712e-05,
      "loss": 0.6639,
      "step": 40
    },
    {
      "epoch": 4.9,
      "learning_rate": 7.5383290016942e-05,
      "loss": 0.6691,
      "step": 41
    },
    {
      "epoch": 5.01,
      "learning_rate": 7.422754351663252e-05,
      "loss": 0.6763,
      "step": 42
    },
    {
      "epoch": 5.01,
      "eval_loss": 0.6810438632965088,
      "eval_runtime": 125.784,
      "eval_samples_per_second": 1.471,
      "eval_steps_per_second": 1.471,
      "step": 42
    },
    {
      "epoch": 5.13,
      "learning_rate": 7.30546250724663e-05,
      "loss": 0.5823,
      "step": 43
    },
    {
      "epoch": 5.25,
      "learning_rate": 7.186536602294278e-05,
      "loss": 0.6099,
      "step": 44
    },
    {
      "epoch": 5.37,
      "learning_rate": 7.066060928841892e-05,
      "loss": 0.619,
      "step": 45
    },
    {
      "epoch": 5.49,
      "learning_rate": 6.944120877366604e-05,
      "loss": 0.649,
      "step": 46
    },
    {
      "epoch": 5.61,
      "learning_rate": 6.820802876264112e-05,
      "loss": 0.589,
      "step": 47
    },
    {
      "epoch": 5.73,
      "learning_rate": 6.696194330590151e-05,
      "loss": 0.6466,
      "step": 48
    },
    {
      "epoch": 5.73,
      "eval_loss": 0.6540355682373047,
      "eval_runtime": 125.4258,
      "eval_samples_per_second": 1.475,
      "eval_steps_per_second": 1.475,
      "step": 48
    },
    {
      "epoch": 5.85,
      "learning_rate": 6.570383560109745e-05,
      "loss": 0.6435,
      "step": 49
    },
    {
      "epoch": 5.97,
      "learning_rate": 6.443459736698105e-05,
      "loss": 0.6231,
      "step": 50
    },
    {
      "epoch": 6.09,
      "learning_rate": 6.315512821137606e-05,
      "loss": 0.6046,
      "step": 51
    },
    {
      "epoch": 6.21,
      "learning_rate": 6.186633499355576e-05,
      "loss": 0.5498,
      "step": 52
    },
    {
      "epoch": 6.33,
      "learning_rate": 6.056913118148122e-05,
      "loss": 0.5987,
      "step": 53
    },
    {
      "epoch": 6.45,
      "learning_rate": 5.9264436204355724e-05,
      "loss": 0.565,
      "step": 54
    },
    {
      "epoch": 6.45,
      "eval_loss": 0.6351470351219177,
      "eval_runtime": 125.5235,
      "eval_samples_per_second": 1.474,
      "eval_steps_per_second": 1.474,
      "step": 54
    },
    {
      "epoch": 6.57,
      "learning_rate": 5.7953174800953604e-05,
      "loss": 0.5997,
      "step": 55
    },
    {
      "epoch": 6.69,
      "learning_rate": 5.6636276364186105e-05,
      "loss": 0.5635,
      "step": 56
    },
    {
      "epoch": 6.81,
      "learning_rate": 5.5314674282368275e-05,
      "loss": 0.5874,
      "step": 57
    },
    {
      "epoch": 6.93,
      "learning_rate": 5.3989305277654156e-05,
      "loss": 0.5615,
      "step": 58
    },
    {
      "epoch": 7.04,
      "learning_rate": 5.2661108742108935e-05,
      "loss": 0.5729,
      "step": 59
    },
    {
      "epoch": 7.16,
      "learning_rate": 5.133102607188874e-05,
      "loss": 0.5736,
      "step": 60
    },
    {
      "epoch": 7.16,
      "eval_loss": 0.6185852885246277,
      "eval_runtime": 125.5344,
      "eval_samples_per_second": 1.474,
      "eval_steps_per_second": 1.474,
      "step": 60
    },
    {
      "epoch": 7.28,
      "learning_rate": 5e-05,
      "loss": 0.5458,
      "step": 61
    },
    {
      "epoch": 7.4,
      "learning_rate": 4.866897392811126e-05,
      "loss": 0.5271,
      "step": 62
    },
    {
      "epoch": 7.52,
      "learning_rate": 4.7338891257891084e-05,
      "loss": 0.5498,
      "step": 63
    },
    {
      "epoch": 7.64,
      "learning_rate": 4.601069472234584e-05,
      "loss": 0.5443,
      "step": 64
    },
    {
      "epoch": 7.76,
      "learning_rate": 4.4685325717631736e-05,
      "loss": 0.5452,
      "step": 65
    },
    {
      "epoch": 7.88,
      "learning_rate": 4.336372363581391e-05,
      "loss": 0.5062,
      "step": 66
    },
    {
      "epoch": 7.88,
      "eval_loss": 0.6089844703674316,
      "eval_runtime": 125.5193,
      "eval_samples_per_second": 1.474,
      "eval_steps_per_second": 1.474,
      "step": 66
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.204682519904641e-05,
      "loss": 0.5397,
      "step": 67
    },
    {
      "epoch": 8.12,
      "learning_rate": 4.0735563795644294e-05,
      "loss": 0.568,
      "step": 68
    },
    {
      "epoch": 8.24,
      "learning_rate": 3.9430868818518784e-05,
      "loss": 0.5134,
      "step": 69
    },
    {
      "epoch": 8.36,
      "learning_rate": 3.8133665006444255e-05,
      "loss": 0.4863,
      "step": 70
    },
    {
      "epoch": 8.48,
      "learning_rate": 3.6844871788623945e-05,
      "loss": 0.5397,
      "step": 71
    },
    {
      "epoch": 8.6,
      "learning_rate": 3.556540263301896e-05,
      "loss": 0.5159,
      "step": 72
    },
    {
      "epoch": 8.6,
      "eval_loss": 0.6018280982971191,
      "eval_runtime": 125.5416,
      "eval_samples_per_second": 1.474,
      "eval_steps_per_second": 1.474,
      "step": 72
    },
    {
      "epoch": 8.72,
      "learning_rate": 3.429616439890258e-05,
      "loss": 0.5079,
      "step": 73
    },
    {
      "epoch": 8.84,
      "learning_rate": 3.303805669409848e-05,
      "loss": 0.4993,
      "step": 74
    },
    {
      "epoch": 8.96,
      "learning_rate": 3.179197123735889e-05,
      "loss": 0.5091,
      "step": 75
    },
    {
      "epoch": 9.07,
      "learning_rate": 3.055879122633397e-05,
      "loss": 0.4833,
      "step": 76
    },
    {
      "epoch": 9.19,
      "learning_rate": 2.9339390711581105e-05,
      "loss": 0.4786,
      "step": 77
    },
    {
      "epoch": 9.31,
      "learning_rate": 2.8134633977057235e-05,
      "loss": 0.5351,
      "step": 78
    },
    {
      "epoch": 9.31,
      "eval_loss": 0.600007176399231,
      "eval_runtime": 125.6028,
      "eval_samples_per_second": 1.473,
      "eval_steps_per_second": 1.473,
      "step": 78
    },
    {
      "epoch": 9.43,
      "learning_rate": 2.69453749275337e-05,
      "loss": 0.5191,
      "step": 79
    },
    {
      "epoch": 9.55,
      "learning_rate": 2.5772456483367497e-05,
      "loss": 0.5325,
      "step": 80
    },
    {
      "epoch": 9.67,
      "learning_rate": 2.4616709983058018e-05,
      "loss": 0.4952,
      "step": 81
    },
    {
      "epoch": 9.79,
      "learning_rate": 2.347895459401288e-05,
      "loss": 0.467,
      "step": 82
    },
    {
      "epoch": 9.91,
      "learning_rate": 2.235999673194035e-05,
      "loss": 0.4942,
      "step": 83
    },
    {
      "epoch": 10.03,
      "learning_rate": 2.126062948927966e-05,
      "loss": 0.4777,
      "step": 84
    },
    {
      "epoch": 10.03,
      "eval_loss": 0.5978866219520569,
      "eval_runtime": 125.5491,
      "eval_samples_per_second": 1.474,
      "eval_steps_per_second": 1.474,
      "step": 84
    },
    {
      "epoch": 10.15,
      "learning_rate": 2.0181632073074926e-05,
      "loss": 0.4731,
      "step": 85
    },
    {
      "epoch": 10.27,
      "learning_rate": 1.912376925269041e-05,
      "loss": 0.5137,
      "step": 86
    },
    {
      "epoch": 10.39,
      "learning_rate": 1.808779081775901e-05,
      "loss": 0.5017,
      "step": 87
    },
    {
      "epoch": 10.51,
      "learning_rate": 1.7074431046748075e-05,
      "loss": 0.468,
      "step": 88
    },
    {
      "epoch": 10.63,
      "learning_rate": 1.6084408186519196e-05,
      "loss": 0.4862,
      "step": 89
    },
    {
      "epoch": 10.75,
      "learning_rate": 1.5118423943250771e-05,
      "loss": 0.5243,
      "step": 90
    },
    {
      "epoch": 10.75,
      "eval_loss": 0.5947938561439514,
      "eval_runtime": 125.4894,
      "eval_samples_per_second": 1.474,
      "eval_steps_per_second": 1.474,
      "step": 90
    },
    {
      "epoch": 10.87,
      "learning_rate": 1.4177162985084242e-05,
      "loss": 0.4662,
      "step": 91
    },
    {
      "epoch": 10.99,
      "learning_rate": 1.3261292456846647e-05,
      "loss": 0.4586,
      "step": 92
    },
    {
      "epoch": 11.1,
      "learning_rate": 1.2371461507193078e-05,
      "loss": 0.4615,
      "step": 93
    },
    {
      "epoch": 11.22,
      "learning_rate": 1.150830082850468e-05,
      "loss": 0.4715,
      "step": 94
    },
    {
      "epoch": 11.34,
      "learning_rate": 1.0672422209867878e-05,
      "loss": 0.5028,
      "step": 95
    },
    {
      "epoch": 11.46,
      "learning_rate": 9.864418103451828e-06,
      "loss": 0.4697,
      "step": 96
    },
    {
      "epoch": 11.46,
      "eval_loss": 0.5938899517059326,
      "eval_runtime": 125.7051,
      "eval_samples_per_second": 1.472,
      "eval_steps_per_second": 1.472,
      "step": 96
    },
    {
      "epoch": 11.58,
      "learning_rate": 9.084861204591549e-06,
      "loss": 0.4815,
      "step": 97
    },
    {
      "epoch": 11.7,
      "learning_rate": 8.334304045874247e-06,
      "loss": 0.482,
      "step": 98
    },
    {
      "epoch": 11.82,
      "learning_rate": 7.613278605516455e-06,
      "loss": 0.5112,
      "step": 99
    },
    {
      "epoch": 11.94,
      "learning_rate": 6.922295930309691e-06,
      "loss": 0.4344,
      "step": 100
    },
    {
      "epoch": 12.06,
      "learning_rate": 6.2618457734019364e-06,
      "loss": 0.4614,
      "step": 101
    },
    {
      "epoch": 12.18,
      "learning_rate": 5.6323962471714286e-06,
      "loss": 0.4704,
      "step": 102
    },
    {
      "epoch": 12.18,
      "eval_loss": 0.5940476655960083,
      "eval_runtime": 125.6268,
      "eval_samples_per_second": 1.473,
      "eval_steps_per_second": 1.473,
      "step": 102
    },
    {
      "epoch": 12.3,
      "learning_rate": 5.034393491439043e-06,
      "loss": 0.4692,
      "step": 103
    },
    {
      "epoch": 12.42,
      "learning_rate": 4.468261357254339e-06,
      "loss": 0.4727,
      "step": 104
    },
    {
      "epoch": 12.54,
      "learning_rate": 3.9344011064793516e-06,
      "loss": 0.4875,
      "step": 105
    },
    {
      "epoch": 12.66,
      "learning_rate": 3.4331911273830784e-06,
      "loss": 0.4759,
      "step": 106
    },
    {
      "epoch": 12.78,
      "learning_rate": 2.9649866664483385e-06,
      "loss": 0.4917,
      "step": 107
    },
    {
      "epoch": 12.9,
      "learning_rate": 2.530119576580936e-06,
      "loss": 0.4917,
      "step": 108
    },
    {
      "epoch": 12.9,
      "eval_loss": 0.5947091579437256,
      "eval_runtime": 125.4985,
      "eval_samples_per_second": 1.474,
      "eval_steps_per_second": 1.474,
      "step": 108
    },
    {
      "epoch": 13.01,
      "learning_rate": 2.1288980818997275e-06,
      "loss": 0.4611,
      "step": 109
    },
    {
      "epoch": 13.13,
      "learning_rate": 1.7616065592742038e-06,
      "loss": 0.4269,
      "step": 110
    },
    {
      "epoch": 13.25,
      "learning_rate": 1.4285053367645074e-06,
      "loss": 0.4576,
      "step": 111
    },
    {
      "epoch": 13.37,
      "learning_rate": 1.1298305091066664e-06,
      "loss": 0.515,
      "step": 112
    },
    {
      "epoch": 13.49,
      "learning_rate": 8.657937703739516e-07,
      "loss": 0.4944,
      "step": 113
    },
    {
      "epoch": 13.61,
      "learning_rate": 6.365822639327723e-07,
      "loss": 0.4665,
      "step": 114
    },
    {
      "epoch": 13.61,
      "eval_loss": 0.594657838344574,
      "eval_runtime": 125.7169,
      "eval_samples_per_second": 1.472,
      "eval_steps_per_second": 1.472,
      "step": 114
    },
    {
      "epoch": 13.73,
      "learning_rate": 4.423584497996458e-07,
      "loss": 0.4911,
      "step": 115
    },
    {
      "epoch": 13.85,
      "learning_rate": 2.8325998949314536e-07,
      "loss": 0.4548,
      "step": 116
    },
    {
      "epoch": 13.97,
      "learning_rate": 1.5939964846249378e-07,
      "loss": 0.4959,
      "step": 117
    },
    {
      "epoch": 14.09,
      "learning_rate": 7.086521616190279e-08,
      "loss": 0.4568,
      "step": 118
    },
    {
      "epoch": 14.21,
      "learning_rate": 1.7719443827368677e-08,
      "loss": 0.4821,
      "step": 119
    },
    {
      "epoch": 14.33,
      "learning_rate": 0.0,
      "loss": 0.4433,
      "step": 120
    },
    {
      "epoch": 14.33,
      "eval_loss": 0.5946800708770752,
      "eval_runtime": 126.38,
      "eval_samples_per_second": 1.464,
      "eval_steps_per_second": 1.464,
      "step": 120
    },
    {
      "epoch": 14.33,
      "step": 120,
      "total_flos": 4.71865270275072e+17,
      "train_loss": 0.6693517704804738,
      "train_runtime": 12311.8994,
      "train_samples_per_second": 0.487,
      "train_steps_per_second": 0.01
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 120,
  "num_train_epochs": 15,
  "save_steps": 12,
  "total_flos": 4.71865270275072e+17,
  "trial_name": null,
  "trial_params": null
}