{
  "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": 134.93724401851352,
      "learning_rate": 4.545454545454545e-08,
      "logits/chosen": 117.67350769042969,
      "logits/rejected": 126.90988159179688,
      "logps/chosen": -336.5020751953125,
      "logps/rejected": -438.0943298339844,
      "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": 136.34386152980758,
      "learning_rate": 4.545454545454545e-07,
      "logits/chosen": 134.7221221923828,
      "logits/rejected": 138.1984100341797,
      "logps/chosen": -395.924560546875,
      "logps/rejected": -439.0712890625,
      "loss": 0.7055,
      "rewards/accuracies": 0.4097222089767456,
      "rewards/chosen": -0.003234411356970668,
      "rewards/margins": -0.03947298228740692,
      "rewards/rejected": 0.036238569766283035,
      "step": 10
    },
    {
      "epoch": 0.3791469194312796,
      "grad_norm": 135.65125923828373,
      "learning_rate": 4.885348141000122e-07,
      "logits/chosen": 121.62422180175781,
      "logits/rejected": 125.15272521972656,
      "logps/chosen": -369.32891845703125,
      "logps/rejected": -422.75689697265625,
      "loss": 0.6228,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": 0.170469269156456,
      "rewards/margins": 0.30520111322402954,
      "rewards/rejected": -0.13473184406757355,
      "step": 20
    },
    {
      "epoch": 0.5687203791469194,
      "grad_norm": 114.07857539812969,
      "learning_rate": 4.5025027361734613e-07,
      "logits/chosen": 141.53656005859375,
      "logits/rejected": 135.02110290527344,
      "logps/chosen": -418.7715759277344,
      "logps/rejected": -464.8955078125,
      "loss": 0.56,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -1.2944904565811157,
      "rewards/margins": 0.8948480486869812,
      "rewards/rejected": -2.189338445663452,
      "step": 30
    },
    {
      "epoch": 0.7582938388625592,
      "grad_norm": 103.49873893671008,
      "learning_rate": 3.893311157806091e-07,
      "logits/chosen": 123.42630767822266,
      "logits/rejected": 112.1140365600586,
      "logps/chosen": -401.87945556640625,
      "logps/rejected": -429.3699645996094,
      "loss": 0.5472,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -2.4228084087371826,
      "rewards/margins": 1.1271207332611084,
      "rewards/rejected": -3.54992938041687,
      "step": 40
    },
    {
      "epoch": 0.9478672985781991,
      "grad_norm": 120.25232045055446,
      "learning_rate": 3.126631330646801e-07,
      "logits/chosen": 138.51483154296875,
      "logits/rejected": 142.50746154785156,
      "logps/chosen": -464.8746643066406,
      "logps/rejected": -547.7520141601562,
      "loss": 0.5034,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -2.2957112789154053,
      "rewards/margins": 1.2795140743255615,
      "rewards/rejected": -3.575225353240967,
      "step": 50
    },
    {
      "epoch": 1.1374407582938388,
      "grad_norm": 60.01877502993502,
      "learning_rate": 2.2891223348923882e-07,
      "logits/chosen": 131.36839294433594,
      "logits/rejected": 134.95950317382812,
      "logps/chosen": -444.547607421875,
      "logps/rejected": -531.6339721679688,
      "loss": 0.3025,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": -2.315918207168579,
      "rewards/margins": 2.439805269241333,
      "rewards/rejected": -4.755723476409912,
      "step": 60
    },
    {
      "epoch": 1.3270142180094786,
      "grad_norm": 47.79216068377628,
      "learning_rate": 1.4754491880085317e-07,
      "logits/chosen": 124.45259094238281,
      "logits/rejected": 126.58074951171875,
      "logps/chosen": -414.2548828125,
      "logps/rejected": -518.2446899414062,
      "loss": 0.1949,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": -2.3579883575439453,
      "rewards/margins": 2.8935599327087402,
      "rewards/rejected": -5.251548767089844,
      "step": 70
    },
    {
      "epoch": 1.5165876777251186,
      "grad_norm": 55.81661982856631,
      "learning_rate": 7.775827023107834e-08,
      "logits/chosen": 110.53846740722656,
      "logits/rejected": 127.3997573852539,
      "logps/chosen": -416.94775390625,
      "logps/rejected": -536.0177001953125,
      "loss": 0.1738,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": -3.1685705184936523,
      "rewards/margins": 2.967393159866333,
      "rewards/rejected": -6.135963439941406,
      "step": 80
    },
    {
      "epoch": 1.7061611374407581,
      "grad_norm": 46.35415773419707,
      "learning_rate": 2.7440387297912122e-08,
      "logits/chosen": 108.8740234375,
      "logits/rejected": 121.63133239746094,
      "logps/chosen": -452.612060546875,
      "logps/rejected": -570.0894775390625,
      "loss": 0.1585,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": -3.32800030708313,
      "rewards/margins": 3.267381191253662,
      "rewards/rejected": -6.595381259918213,
      "step": 90
    },
    {
      "epoch": 1.8957345971563981,
      "grad_norm": 43.55253667946353,
      "learning_rate": 2.27878296044029e-09,
      "logits/chosen": 115.157958984375,
      "logits/rejected": 114.7303695678711,
      "logps/chosen": -443.5245056152344,
      "logps/rejected": -538.7801513671875,
      "loss": 0.1534,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -3.0830078125,
      "rewards/margins": 3.0285675525665283,
      "rewards/rejected": -6.111575126647949,
      "step": 100
    },
    {
      "epoch": 1.8957345971563981,
      "eval_logits/chosen": 94.33853912353516,
      "eval_logits/rejected": 88.46772003173828,
      "eval_logps/chosen": -442.3636779785156,
      "eval_logps/rejected": -472.6589660644531,
      "eval_loss": 0.4630958139896393,
      "eval_rewards/accuracies": 0.7083333134651184,
      "eval_rewards/chosen": -3.9814913272857666,
      "eval_rewards/margins": 1.5538396835327148,
      "eval_rewards/rejected": -5.5353312492370605,
      "eval_runtime": 21.3945,
      "eval_samples_per_second": 35.056,
      "eval_steps_per_second": 1.122,
      "step": 100
    },
    {
      "epoch": 1.971563981042654,
      "step": 104,
      "total_flos": 0.0,
      "train_loss": 0.38818228932527393,
      "train_runtime": 1555.646,
      "train_samples_per_second": 8.678,
      "train_steps_per_second": 0.067
    }
  ],
  "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
}