{
  "best_metric": 0.9180520176887512,
  "best_model_checkpoint": "ckpt/origin/pedes_attention_v2/checkpoint-91",
  "epoch": 4.882629107981221,
  "eval_steps": 7,
  "global_step": 130,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 5e-06,
      "loss": 1.0282,
      "step": 1
    },
    {
      "epoch": 0.08,
      "learning_rate": 1e-05,
      "loss": 0.9767,
      "step": 2
    },
    {
      "epoch": 0.11,
      "learning_rate": 9.998494093481022e-06,
      "loss": 1.0075,
      "step": 3
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.993977281025862e-06,
      "loss": 0.9679,
      "step": 4
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.986452283393452e-06,
      "loss": 0.971,
      "step": 5
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.975923633360985e-06,
      "loss": 1.0506,
      "step": 6
    },
    {
      "epoch": 0.26,
      "learning_rate": 9.962397672993552e-06,
      "loss": 1.0311,
      "step": 7
    },
    {
      "epoch": 0.26,
      "eval_loss": 0.9760443568229675,
      "eval_runtime": 37.1633,
      "eval_samples_per_second": 1.91,
      "eval_steps_per_second": 1.91,
      "step": 7
    },
    {
      "epoch": 0.3,
      "learning_rate": 9.945882549823906e-06,
      "loss": 1.0691,
      "step": 8
    },
    {
      "epoch": 0.34,
      "learning_rate": 9.926388211944707e-06,
      "loss": 0.9659,
      "step": 9
    },
    {
      "epoch": 0.38,
      "learning_rate": 9.903926402016153e-06,
      "loss": 1.0062,
      "step": 10
    },
    {
      "epoch": 0.41,
      "learning_rate": 9.878510650192644e-06,
      "loss": 1.0148,
      "step": 11
    },
    {
      "epoch": 0.45,
      "learning_rate": 9.850156265972722e-06,
      "loss": 0.9223,
      "step": 12
    },
    {
      "epoch": 0.49,
      "learning_rate": 9.8188803289772e-06,
      "loss": 1.0146,
      "step": 13
    },
    {
      "epoch": 0.53,
      "learning_rate": 9.784701678661045e-06,
      "loss": 0.97,
      "step": 14
    },
    {
      "epoch": 0.53,
      "eval_loss": 0.9596567153930664,
      "eval_runtime": 36.9759,
      "eval_samples_per_second": 1.92,
      "eval_steps_per_second": 1.92,
      "step": 14
    },
    {
      "epoch": 0.56,
      "learning_rate": 9.747640902965185e-06,
      "loss": 1.0113,
      "step": 15
    },
    {
      "epoch": 0.6,
      "learning_rate": 9.707720325915105e-06,
      "loss": 0.9954,
      "step": 16
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.664963994173695e-06,
      "loss": 1.0285,
      "step": 17
    },
    {
      "epoch": 0.68,
      "learning_rate": 9.619397662556434e-06,
      "loss": 1.042,
      "step": 18
    },
    {
      "epoch": 0.71,
      "learning_rate": 9.571048778517655e-06,
      "loss": 0.9092,
      "step": 19
    },
    {
      "epoch": 0.75,
      "learning_rate": 9.519946465617217e-06,
      "loss": 0.932,
      "step": 20
    },
    {
      "epoch": 0.79,
      "learning_rate": 9.466121505977577e-06,
      "loss": 1.014,
      "step": 21
    },
    {
      "epoch": 0.79,
      "eval_loss": 0.9468032121658325,
      "eval_runtime": 36.927,
      "eval_samples_per_second": 1.923,
      "eval_steps_per_second": 1.923,
      "step": 21
    },
    {
      "epoch": 0.83,
      "learning_rate": 9.409606321741776e-06,
      "loss": 1.0309,
      "step": 22
    },
    {
      "epoch": 0.86,
      "learning_rate": 9.350434955543557e-06,
      "loss": 0.9213,
      "step": 23
    },
    {
      "epoch": 0.9,
      "learning_rate": 9.288643050001362e-06,
      "loss": 0.9605,
      "step": 24
    },
    {
      "epoch": 0.94,
      "learning_rate": 9.224267826248536e-06,
      "loss": 0.9896,
      "step": 25
    },
    {
      "epoch": 0.98,
      "learning_rate": 9.157348061512728e-06,
      "loss": 0.9815,
      "step": 26
    },
    {
      "epoch": 1.01,
      "learning_rate": 9.08792406575792e-06,
      "loss": 0.9748,
      "step": 27
    },
    {
      "epoch": 1.05,
      "learning_rate": 9.016037657403225e-06,
      "loss": 0.993,
      "step": 28
    },
    {
      "epoch": 1.05,
      "eval_loss": 0.9380103945732117,
      "eval_runtime": 36.9631,
      "eval_samples_per_second": 1.921,
      "eval_steps_per_second": 1.921,
      "step": 28
    },
    {
      "epoch": 1.09,
      "learning_rate": 8.941732138133032e-06,
      "loss": 0.9235,
      "step": 29
    },
    {
      "epoch": 1.13,
      "learning_rate": 8.865052266813686e-06,
      "loss": 0.9419,
      "step": 30
    },
    {
      "epoch": 1.16,
      "learning_rate": 8.786044232532423e-06,
      "loss": 0.9539,
      "step": 31
    },
    {
      "epoch": 1.2,
      "learning_rate": 8.704755626774796e-06,
      "loss": 1.016,
      "step": 32
    },
    {
      "epoch": 1.24,
      "learning_rate": 8.621235414757337e-06,
      "loss": 0.9607,
      "step": 33
    },
    {
      "epoch": 1.28,
      "learning_rate": 8.535533905932739e-06,
      "loss": 0.9749,
      "step": 34
    },
    {
      "epoch": 1.31,
      "learning_rate": 8.447702723685335e-06,
      "loss": 0.9689,
      "step": 35
    },
    {
      "epoch": 1.31,
      "eval_loss": 0.933106005191803,
      "eval_runtime": 37.0437,
      "eval_samples_per_second": 1.917,
      "eval_steps_per_second": 1.917,
      "step": 35
    },
    {
      "epoch": 1.35,
      "learning_rate": 8.357794774235094e-06,
      "loss": 0.9367,
      "step": 36
    },
    {
      "epoch": 1.39,
      "learning_rate": 8.265864214768883e-06,
      "loss": 0.8935,
      "step": 37
    },
    {
      "epoch": 1.43,
      "learning_rate": 8.171966420818227e-06,
      "loss": 0.9468,
      "step": 38
    },
    {
      "epoch": 1.46,
      "learning_rate": 8.076157952903134e-06,
      "loss": 0.9914,
      "step": 39
    },
    {
      "epoch": 1.5,
      "learning_rate": 7.978496522462167e-06,
      "loss": 0.9485,
      "step": 40
    },
    {
      "epoch": 1.54,
      "learning_rate": 7.879040957089229e-06,
      "loss": 0.9789,
      "step": 41
    },
    {
      "epoch": 1.58,
      "learning_rate": 7.777851165098012e-06,
      "loss": 0.9744,
      "step": 42
    },
    {
      "epoch": 1.58,
      "eval_loss": 0.9296567440032959,
      "eval_runtime": 37.085,
      "eval_samples_per_second": 1.915,
      "eval_steps_per_second": 1.915,
      "step": 42
    },
    {
      "epoch": 1.62,
      "learning_rate": 7.674988099435487e-06,
      "loss": 1.0782,
      "step": 43
    },
    {
      "epoch": 1.65,
      "learning_rate": 7.570513720966108e-06,
      "loss": 0.997,
      "step": 44
    },
    {
      "epoch": 1.69,
      "learning_rate": 7.464490961148921e-06,
      "loss": 0.9033,
      "step": 45
    },
    {
      "epoch": 1.73,
      "learning_rate": 7.3569836841299905e-06,
      "loss": 0.8924,
      "step": 46
    },
    {
      "epoch": 1.77,
      "learning_rate": 7.248056648273034e-06,
      "loss": 0.9623,
      "step": 47
    },
    {
      "epoch": 1.8,
      "learning_rate": 7.137775467151411e-06,
      "loss": 0.922,
      "step": 48
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.026206570024949e-06,
      "loss": 0.9452,
      "step": 49
    },
    {
      "epoch": 1.84,
      "eval_loss": 0.9257412552833557,
      "eval_runtime": 37.1517,
      "eval_samples_per_second": 1.911,
      "eval_steps_per_second": 1.911,
      "step": 49
    },
    {
      "epoch": 1.88,
      "learning_rate": 6.913417161825449e-06,
      "loss": 0.9623,
      "step": 50
    },
    {
      "epoch": 1.92,
      "learning_rate": 6.799475182674942e-06,
      "loss": 0.8907,
      "step": 51
    },
    {
      "epoch": 1.95,
      "learning_rate": 6.684449266961101e-06,
      "loss": 0.8867,
      "step": 52
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.568408701994459e-06,
      "loss": 0.9887,
      "step": 53
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.451423386272312e-06,
      "loss": 0.9524,
      "step": 54
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.333563787374493e-06,
      "loss": 0.8912,
      "step": 55
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.21490089951632e-06,
      "loss": 0.9499,
      "step": 56
    },
    {
      "epoch": 2.1,
      "eval_loss": 0.9232458472251892,
      "eval_runtime": 37.0638,
      "eval_samples_per_second": 1.916,
      "eval_steps_per_second": 1.916,
      "step": 56
    },
    {
      "epoch": 2.14,
      "learning_rate": 6.095506200784349e-06,
      "loss": 0.8558,
      "step": 57
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.975451610080643e-06,
      "loss": 0.968,
      "step": 58
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.8548094438015065e-06,
      "loss": 0.9154,
      "step": 59
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.733652372276809e-06,
      "loss": 0.9611,
      "step": 60
    },
    {
      "epoch": 2.29,
      "learning_rate": 5.612053375996082e-06,
      "loss": 0.9836,
      "step": 61
    },
    {
      "epoch": 2.33,
      "learning_rate": 5.490085701647805e-06,
      "loss": 0.9307,
      "step": 62
    },
    {
      "epoch": 2.37,
      "learning_rate": 5.367822817998338e-06,
      "loss": 1.0362,
      "step": 63
    },
    {
      "epoch": 2.37,
      "eval_loss": 0.9205808043479919,
      "eval_runtime": 37.0451,
      "eval_samples_per_second": 1.917,
      "eval_steps_per_second": 1.917,
      "step": 63
    },
    {
      "epoch": 2.4,
      "learning_rate": 5.245338371637091e-06,
      "loss": 0.9823,
      "step": 64
    },
    {
      "epoch": 2.44,
      "learning_rate": 5.122706142614562e-06,
      "loss": 0.9406,
      "step": 65
    },
    {
      "epoch": 2.48,
      "learning_rate": 5e-06,
      "loss": 0.9989,
      "step": 66
    },
    {
      "epoch": 2.52,
      "learning_rate": 4.87729385738544e-06,
      "loss": 0.8766,
      "step": 67
    },
    {
      "epoch": 2.55,
      "learning_rate": 4.75466162836291e-06,
      "loss": 0.9258,
      "step": 68
    },
    {
      "epoch": 2.59,
      "learning_rate": 4.6321771820016635e-06,
      "loss": 0.9525,
      "step": 69
    },
    {
      "epoch": 2.63,
      "learning_rate": 4.509914298352197e-06,
      "loss": 0.9574,
      "step": 70
    },
    {
      "epoch": 2.63,
      "eval_loss": 0.9198555946350098,
      "eval_runtime": 36.934,
      "eval_samples_per_second": 1.922,
      "eval_steps_per_second": 1.922,
      "step": 70
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.38794662400392e-06,
      "loss": 0.9648,
      "step": 71
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.266347627723192e-06,
      "loss": 0.9437,
      "step": 72
    },
    {
      "epoch": 2.74,
      "learning_rate": 4.145190556198494e-06,
      "loss": 0.8811,
      "step": 73
    },
    {
      "epoch": 2.78,
      "learning_rate": 4.02454838991936e-06,
      "loss": 0.9659,
      "step": 74
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.904493799215652e-06,
      "loss": 0.9659,
      "step": 75
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.7850991004836813e-06,
      "loss": 0.8528,
      "step": 76
    },
    {
      "epoch": 2.89,
      "learning_rate": 3.6664362126255087e-06,
      "loss": 0.9295,
      "step": 77
    },
    {
      "epoch": 2.89,
      "eval_loss": 0.9187451004981995,
      "eval_runtime": 37.0034,
      "eval_samples_per_second": 1.919,
      "eval_steps_per_second": 1.919,
      "step": 77
    },
    {
      "epoch": 2.93,
      "learning_rate": 3.5485766137276894e-06,
      "loss": 0.9386,
      "step": 78
    },
    {
      "epoch": 2.97,
      "learning_rate": 3.4315912980055433e-06,
      "loss": 0.9061,
      "step": 79
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.3155507330389004e-06,
      "loss": 0.9602,
      "step": 80
    },
    {
      "epoch": 3.04,
      "learning_rate": 3.2005248173250593e-06,
      "loss": 0.9184,
      "step": 81
    },
    {
      "epoch": 3.08,
      "learning_rate": 3.0865828381745515e-06,
      "loss": 1.0052,
      "step": 82
    },
    {
      "epoch": 3.12,
      "learning_rate": 2.9737934299750514e-06,
      "loss": 0.9061,
      "step": 83
    },
    {
      "epoch": 3.15,
      "learning_rate": 2.862224532848591e-06,
      "loss": 0.9485,
      "step": 84
    },
    {
      "epoch": 3.15,
      "eval_loss": 0.9179951548576355,
      "eval_runtime": 36.9583,
      "eval_samples_per_second": 1.921,
      "eval_steps_per_second": 1.921,
      "step": 84
    },
    {
      "epoch": 3.19,
      "learning_rate": 2.7519433517269665e-06,
      "loss": 0.9743,
      "step": 85
    },
    {
      "epoch": 3.23,
      "learning_rate": 2.6430163158700116e-06,
      "loss": 0.9124,
      "step": 86
    },
    {
      "epoch": 3.27,
      "learning_rate": 2.5355090388510806e-06,
      "loss": 0.9506,
      "step": 87
    },
    {
      "epoch": 3.31,
      "learning_rate": 2.429486279033892e-06,
      "loss": 0.9314,
      "step": 88
    },
    {
      "epoch": 3.34,
      "learning_rate": 2.325011900564515e-06,
      "loss": 0.9229,
      "step": 89
    },
    {
      "epoch": 3.38,
      "learning_rate": 2.2221488349019903e-06,
      "loss": 0.9828,
      "step": 90
    },
    {
      "epoch": 3.42,
      "learning_rate": 2.1209590429107734e-06,
      "loss": 0.9736,
      "step": 91
    },
    {
      "epoch": 3.42,
      "eval_loss": 0.9180520176887512,
      "eval_runtime": 36.9028,
      "eval_samples_per_second": 1.924,
      "eval_steps_per_second": 1.924,
      "step": 91
    },
    {
      "epoch": 3.46,
      "learning_rate": 2.0215034775378336e-06,
      "loss": 1.0039,
      "step": 92
    },
    {
      "epoch": 3.49,
      "learning_rate": 1.9238420470968665e-06,
      "loss": 0.9363,
      "step": 93
    },
    {
      "epoch": 3.53,
      "learning_rate": 1.8280335791817733e-06,
      "loss": 0.9026,
      "step": 94
    },
    {
      "epoch": 3.57,
      "learning_rate": 1.7341357852311175e-06,
      "loss": 1.0065,
      "step": 95
    },
    {
      "epoch": 3.61,
      "learning_rate": 1.642205225764908e-06,
      "loss": 0.9446,
      "step": 96
    },
    {
      "epoch": 3.64,
      "learning_rate": 1.5522972763146653e-06,
      "loss": 0.9325,
      "step": 97
    },
    {
      "epoch": 3.68,
      "learning_rate": 1.4644660940672628e-06,
      "loss": 0.904,
      "step": 98
    },
    {
      "epoch": 3.68,
      "eval_loss": 0.9174049496650696,
      "eval_runtime": 36.9039,
      "eval_samples_per_second": 1.924,
      "eval_steps_per_second": 1.924,
      "step": 98
    },
    {
      "epoch": 3.72,
      "learning_rate": 1.3787645852426663e-06,
      "loss": 0.8823,
      "step": 99
    },
    {
      "epoch": 3.76,
      "learning_rate": 1.2952443732252058e-06,
      "loss": 0.9824,
      "step": 100
    },
    {
      "epoch": 3.79,
      "learning_rate": 1.2139557674675773e-06,
      "loss": 0.884,
      "step": 101
    },
    {
      "epoch": 3.83,
      "learning_rate": 1.134947733186315e-06,
      "loss": 0.8807,
      "step": 102
    },
    {
      "epoch": 3.87,
      "learning_rate": 1.058267861866969e-06,
      "loss": 0.9713,
      "step": 103
    },
    {
      "epoch": 3.91,
      "learning_rate": 9.83962342596776e-07,
      "loss": 0.9151,
      "step": 104
    },
    {
      "epoch": 3.94,
      "learning_rate": 9.120759342420821e-07,
      "loss": 0.9174,
      "step": 105
    },
    {
      "epoch": 3.94,
      "eval_loss": 0.917622447013855,
      "eval_runtime": 36.8966,
      "eval_samples_per_second": 1.924,
      "eval_steps_per_second": 1.924,
      "step": 105
    },
    {
      "epoch": 3.98,
      "learning_rate": 8.426519384872733e-07,
      "loss": 0.8545,
      "step": 106
    },
    {
      "epoch": 4.02,
      "learning_rate": 7.757321737514645e-07,
      "loss": 0.9481,
      "step": 107
    },
    {
      "epoch": 4.06,
      "learning_rate": 7.113569499986401e-07,
      "loss": 0.9625,
      "step": 108
    },
    {
      "epoch": 4.09,
      "learning_rate": 6.495650444564433e-07,
      "loss": 0.9481,
      "step": 109
    },
    {
      "epoch": 4.13,
      "learning_rate": 5.903936782582253e-07,
      "loss": 0.9439,
      "step": 110
    },
    {
      "epoch": 4.17,
      "learning_rate": 5.338784940224239e-07,
      "loss": 0.902,
      "step": 111
    },
    {
      "epoch": 4.21,
      "learning_rate": 4.800535343827834e-07,
      "loss": 0.9468,
      "step": 112
    },
    {
      "epoch": 4.21,
      "eval_loss": 0.9171994924545288,
      "eval_runtime": 37.0036,
      "eval_samples_per_second": 1.919,
      "eval_steps_per_second": 1.919,
      "step": 112
    },
    {
      "epoch": 4.24,
      "learning_rate": 4.289512214823466e-07,
      "loss": 0.9181,
      "step": 113
    },
    {
      "epoch": 4.28,
      "learning_rate": 3.8060233744356634e-07,
      "loss": 0.9299,
      "step": 114
    },
    {
      "epoch": 4.32,
      "learning_rate": 3.350360058263058e-07,
      "loss": 0.8488,
      "step": 115
    },
    {
      "epoch": 4.36,
      "learning_rate": 2.9227967408489653e-07,
      "loss": 0.894,
      "step": 116
    },
    {
      "epoch": 4.39,
      "learning_rate": 2.523590970348166e-07,
      "loss": 0.9252,
      "step": 117
    },
    {
      "epoch": 4.43,
      "learning_rate": 2.152983213389559e-07,
      "loss": 0.9887,
      "step": 118
    },
    {
      "epoch": 4.47,
      "learning_rate": 1.8111967102280082e-07,
      "loss": 0.9491,
      "step": 119
    },
    {
      "epoch": 4.47,
      "eval_loss": 0.9177583456039429,
      "eval_runtime": 36.857,
      "eval_samples_per_second": 1.926,
      "eval_steps_per_second": 1.926,
      "step": 119
    },
    {
      "epoch": 4.51,
      "learning_rate": 1.4984373402728014e-07,
      "loss": 0.9271,
      "step": 120
    },
    {
      "epoch": 4.54,
      "learning_rate": 1.2148934980735772e-07,
      "loss": 0.9753,
      "step": 121
    },
    {
      "epoch": 4.58,
      "learning_rate": 9.607359798384785e-08,
      "loss": 0.9216,
      "step": 122
    },
    {
      "epoch": 4.62,
      "learning_rate": 7.36117880552939e-08,
      "loss": 0.8974,
      "step": 123
    },
    {
      "epoch": 4.66,
      "learning_rate": 5.411745017609493e-08,
      "loss": 0.9348,
      "step": 124
    },
    {
      "epoch": 4.69,
      "learning_rate": 3.7602327006450166e-08,
      "loss": 0.9237,
      "step": 125
    },
    {
      "epoch": 4.73,
      "learning_rate": 2.4076366639015914e-08,
      "loss": 0.9956,
      "step": 126
    },
    {
      "epoch": 4.73,
      "eval_loss": 0.9172103404998779,
      "eval_runtime": 36.9828,
      "eval_samples_per_second": 1.92,
      "eval_steps_per_second": 1.92,
      "step": 126
    },
    {
      "epoch": 4.77,
      "learning_rate": 1.3547716606548967e-08,
      "loss": 0.8845,
      "step": 127
    },
    {
      "epoch": 4.81,
      "learning_rate": 6.022718974137976e-09,
      "loss": 0.8938,
      "step": 128
    },
    {
      "epoch": 4.85,
      "learning_rate": 1.5059065189787502e-09,
      "loss": 0.931,
      "step": 129
    },
    {
      "epoch": 4.88,
      "learning_rate": 0.0,
      "loss": 0.9201,
      "step": 130
    },
    {
      "epoch": 4.88,
      "step": 130,
      "total_flos": 3.5067525347672064e+17,
      "train_loss": 0.9512384671431321,
      "train_runtime": 7720.1122,
      "train_samples_per_second": 0.828,
      "train_steps_per_second": 0.017
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 130,
  "num_train_epochs": 5,
  "save_steps": 13,
  "total_flos": 3.5067525347672064e+17,
  "trial_name": null,
  "trial_params": null
}