{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.971563981042654,
  "eval_steps": 100,
  "global_step": 104,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.018957345971563982,
      "grad_norm": 132.20305622646663,
      "learning_rate": 4.545454545454545e-08,
      "logits/chosen": 117.53560638427734,
      "logits/rejected": 126.8960952758789,
      "logps/chosen": -335.40118408203125,
      "logps/rejected": -439.16552734375,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.1895734597156398,
      "grad_norm": 132.5788441374347,
      "learning_rate": 4.545454545454545e-07,
      "logits/chosen": 135.00689697265625,
      "logits/rejected": 138.35719299316406,
      "logps/chosen": -396.1441955566406,
      "logps/rejected": -439.40240478515625,
      "loss": 0.7179,
      "rewards/accuracies": 0.4444444477558136,
      "rewards/chosen": -0.007382770534604788,
      "rewards/margins": -0.0036396842915564775,
      "rewards/rejected": -0.003743087174370885,
      "step": 10
    },
    {
      "epoch": 0.3791469194312796,
      "grad_norm": 130.1388638934296,
      "learning_rate": 4.885348141000122e-07,
      "logits/chosen": 121.08137512207031,
      "logits/rejected": 124.82417297363281,
      "logps/chosen": -371.0042419433594,
      "logps/rejected": -424.25299072265625,
      "loss": 0.6433,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": 0.07038744539022446,
      "rewards/margins": 0.28346824645996094,
      "rewards/rejected": -0.21308080852031708,
      "step": 20
    },
    {
      "epoch": 0.5687203791469194,
      "grad_norm": 119.85468134756464,
      "learning_rate": 4.5025027361734613e-07,
      "logits/chosen": 142.1728057861328,
      "logits/rejected": 135.75143432617188,
      "logps/chosen": -421.06903076171875,
      "logps/rejected": -467.2119140625,
      "loss": 0.5746,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -1.4301902055740356,
      "rewards/margins": 0.9302951693534851,
      "rewards/rejected": -2.360485553741455,
      "step": 30
    },
    {
      "epoch": 0.7582938388625592,
      "grad_norm": 108.85906420456956,
      "learning_rate": 3.893311157806091e-07,
      "logits/chosen": 125.63570404052734,
      "logits/rejected": 113.9980697631836,
      "logps/chosen": -397.3561706542969,
      "logps/rejected": -424.4190368652344,
      "loss": 0.5496,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": -2.158099412918091,
      "rewards/margins": 1.169090986251831,
      "rewards/rejected": -3.327190399169922,
      "step": 40
    },
    {
      "epoch": 0.9478672985781991,
      "grad_norm": 121.38249679586032,
      "learning_rate": 3.126631330646801e-07,
      "logits/chosen": 140.33822631835938,
      "logits/rejected": 144.39553833007812,
      "logps/chosen": -462.13653564453125,
      "logps/rejected": -546.0151977539062,
      "loss": 0.4858,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -2.1496312618255615,
      "rewards/margins": 1.3348098993301392,
      "rewards/rejected": -3.484441041946411,
      "step": 50
    },
    {
      "epoch": 1.1374407582938388,
      "grad_norm": 60.277332590993936,
      "learning_rate": 2.2891223348923882e-07,
      "logits/chosen": 131.23561096191406,
      "logits/rejected": 134.5987091064453,
      "logps/chosen": -446.07977294921875,
      "logps/rejected": -534.3648071289062,
      "loss": 0.3021,
      "rewards/accuracies": 0.887499988079071,
      "rewards/chosen": -2.403707265853882,
      "rewards/margins": 2.4962546825408936,
      "rewards/rejected": -4.899962425231934,
      "step": 60
    },
    {
      "epoch": 1.3270142180094786,
      "grad_norm": 41.03323290379581,
      "learning_rate": 1.4754491880085317e-07,
      "logits/chosen": 121.9046401977539,
      "logits/rejected": 124.14168548583984,
      "logps/chosen": -419.36175537109375,
      "logps/rejected": -526.709228515625,
      "loss": 0.1837,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -2.6108951568603516,
      "rewards/margins": 3.0488715171813965,
      "rewards/rejected": -5.659766674041748,
      "step": 70
    },
    {
      "epoch": 1.5165876777251186,
      "grad_norm": 59.303132679746895,
      "learning_rate": 7.775827023107834e-08,
      "logits/chosen": 107.08919525146484,
      "logits/rejected": 124.07452392578125,
      "logps/chosen": -425.52716064453125,
      "logps/rejected": -548.1698608398438,
      "loss": 0.1693,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": -3.5938053131103516,
      "rewards/margins": 3.152803897857666,
      "rewards/rejected": -6.746609687805176,
      "step": 80
    },
    {
      "epoch": 1.7061611374407581,
      "grad_norm": 49.07574340329523,
      "learning_rate": 2.7440387297912122e-08,
      "logits/chosen": 105.5522232055664,
      "logits/rejected": 118.3895492553711,
      "logps/chosen": -459.85089111328125,
      "logps/rejected": -583.5906982421875,
      "loss": 0.1517,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": -3.670989513397217,
      "rewards/margins": 3.62445330619812,
      "rewards/rejected": -7.295443058013916,
      "step": 90
    },
    {
      "epoch": 1.8957345971563981,
      "grad_norm": 48.77979045218656,
      "learning_rate": 2.27878296044029e-09,
      "logits/chosen": 111.78726959228516,
      "logits/rejected": 111.61686706542969,
      "logps/chosen": -448.10687255859375,
      "logps/rejected": -548.4755859375,
      "loss": 0.1497,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": -3.3065879344940186,
      "rewards/margins": 3.278989315032959,
      "rewards/rejected": -6.585577964782715,
      "step": 100
    },
    {
      "epoch": 1.8957345971563981,
      "eval_logits/chosen": 91.48928833007812,
      "eval_logits/rejected": 85.78076934814453,
      "eval_logps/chosen": -446.239990234375,
      "eval_logps/rejected": -478.9853515625,
      "eval_loss": 0.460786372423172,
      "eval_rewards/accuracies": 0.7291666865348816,
      "eval_rewards/chosen": -4.128589153289795,
      "eval_rewards/margins": 1.7171281576156616,
      "eval_rewards/rejected": -5.845716953277588,
      "eval_runtime": 128.1398,
      "eval_samples_per_second": 5.853,
      "eval_steps_per_second": 0.187,
      "step": 100
    },
    {
      "epoch": 1.971563981042654,
      "step": 104,
      "total_flos": 0.0,
      "train_loss": 0.38906457561713,
      "train_runtime": 3484.4936,
      "train_samples_per_second": 3.874,
      "train_steps_per_second": 0.03
    }
  ],
  "logging_steps": 10,
  "max_steps": 104,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}