{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.23529411764705882,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0,
      "eval_logits/chosen": -2.3223893642425537,
      "eval_logits/rejected": -1.7441076040267944,
      "eval_logps/chosen": -205.81431579589844,
      "eval_logps/rejected": -160.84881591796875,
      "eval_loss": 0.6931471824645996,
      "eval_rewards/accuracies": 0.0,
      "eval_rewards/chosen": 0.0,
      "eval_rewards/margins": 0.0,
      "eval_rewards/rejected": 0.0,
      "eval_runtime": 13.1039,
      "eval_samples_per_second": 1.297,
      "eval_steps_per_second": 0.687,
      "step": 0
    },
    {
      "epoch": 0.026143790849673203,
      "grad_norm": 11.304482460021973,
      "learning_rate": 0.0,
      "logits/chosen": -2.2234325408935547,
      "logits/rejected": -2.2330026626586914,
      "logps/chosen": -251.4762420654297,
      "logps/rejected": -287.5872497558594,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.05228758169934641,
      "grad_norm": 11.91282844543457,
      "learning_rate": 2e-05,
      "logits/chosen": -2.2173714637756348,
      "logits/rejected": -2.2363157272338867,
      "logps/chosen": -227.2211151123047,
      "logps/rejected": -262.26251220703125,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 2
    },
    {
      "epoch": 0.0784313725490196,
      "grad_norm": 9.456698417663574,
      "learning_rate": 4e-05,
      "logits/chosen": -1.7777986526489258,
      "logits/rejected": -1.5732976198196411,
      "logps/chosen": -166.3558807373047,
      "logps/rejected": -206.31166076660156,
      "loss": 0.6976,
      "rewards/accuracies": 0.375,
      "rewards/chosen": 0.005008268635720015,
      "rewards/margins": -0.007055856287479401,
      "rewards/rejected": 0.012064123526215553,
      "step": 3
    },
    {
      "epoch": 0.0784313725490196,
      "eval_logits/chosen": -2.3259360790252686,
      "eval_logits/rejected": -1.7466574907302856,
      "eval_logps/chosen": -205.6438751220703,
      "eval_logps/rejected": -160.8700408935547,
      "eval_loss": 0.6828210353851318,
      "eval_rewards/accuracies": 0.6666666865348816,
      "eval_rewards/chosen": 0.017044195905327797,
      "eval_rewards/margins": 0.019166946411132812,
      "eval_rewards/rejected": -0.0021227519027888775,
      "eval_runtime": 12.9552,
      "eval_samples_per_second": 1.312,
      "eval_steps_per_second": 0.695,
      "step": 3
    },
    {
      "epoch": 0.10457516339869281,
      "grad_norm": 11.230616569519043,
      "learning_rate": 6e-05,
      "logits/chosen": -2.7505903244018555,
      "logits/rejected": -1.348630428314209,
      "logps/chosen": -209.1761474609375,
      "logps/rejected": -190.92935180664062,
      "loss": 0.6807,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -0.005297848954796791,
      "rewards/margins": 0.027892589569091797,
      "rewards/rejected": -0.033190444111824036,
      "step": 4
    },
    {
      "epoch": 0.13071895424836602,
      "grad_norm": 9.93552017211914,
      "learning_rate": 8e-05,
      "logits/chosen": -1.8584319353103638,
      "logits/rejected": -1.2940995693206787,
      "logps/chosen": -245.94874572753906,
      "logps/rejected": -135.0790252685547,
      "loss": 0.6947,
      "rewards/accuracies": 0.5,
      "rewards/chosen": -0.0011208057403564453,
      "rewards/margins": 0.0006671445444226265,
      "rewards/rejected": -0.0017879479564726353,
      "step": 5
    },
    {
      "epoch": 0.1568627450980392,
      "grad_norm": 11.156682014465332,
      "learning_rate": 0.0001,
      "logits/chosen": -2.436879873275757,
      "logits/rejected": -2.5771214962005615,
      "logps/chosen": -204.76026916503906,
      "logps/rejected": -262.71990966796875,
      "loss": 0.6572,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -0.015309764072299004,
      "rewards/margins": 0.07809243351221085,
      "rewards/rejected": -0.093402199447155,
      "step": 6
    },
    {
      "epoch": 0.1568627450980392,
      "eval_logits/chosen": -2.3328909873962402,
      "eval_logits/rejected": -1.7542129755020142,
      "eval_logps/chosen": -205.6370391845703,
      "eval_logps/rejected": -161.59091186523438,
      "eval_loss": 0.6492722034454346,
      "eval_rewards/accuracies": 0.6666666865348816,
      "eval_rewards/chosen": 0.017727533355355263,
      "eval_rewards/margins": 0.09193826466798782,
      "eval_rewards/rejected": -0.07421071827411652,
      "eval_runtime": 11.3826,
      "eval_samples_per_second": 1.494,
      "eval_steps_per_second": 0.791,
      "step": 6
    },
    {
      "epoch": 0.1830065359477124,
      "grad_norm": 9.182978630065918,
      "learning_rate": 0.00012,
      "logits/chosen": -3.110889434814453,
      "logits/rejected": -2.425046443939209,
      "logps/chosen": -183.1647186279297,
      "logps/rejected": -238.2554168701172,
      "loss": 0.6602,
      "rewards/accuracies": 0.625,
      "rewards/chosen": 0.02616577222943306,
      "rewards/margins": 0.07209749519824982,
      "rewards/rejected": -0.04593172296881676,
      "step": 7
    },
    {
      "epoch": 0.20915032679738563,
      "grad_norm": 10.583196640014648,
      "learning_rate": 0.00014,
      "logits/chosen": -2.9872114658355713,
      "logits/rejected": -2.2816989421844482,
      "logps/chosen": -195.46536254882812,
      "logps/rejected": -200.18411254882812,
      "loss": 0.6859,
      "rewards/accuracies": 0.5,
      "rewards/chosen": -0.03734993934631348,
      "rewards/margins": 0.020749665796756744,
      "rewards/rejected": -0.05809960514307022,
      "step": 8
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 8.478883743286133,
      "learning_rate": 0.00016,
      "logits/chosen": -2.448885440826416,
      "logits/rejected": -1.7898311614990234,
      "logps/chosen": -121.42498779296875,
      "logps/rejected": -155.28976440429688,
      "loss": 0.6212,
      "rewards/accuracies": 0.625,
      "rewards/chosen": 0.11882638931274414,
      "rewards/margins": 0.1818961203098297,
      "rewards/rejected": -0.06306972354650497,
      "step": 9
    },
    {
      "epoch": 0.23529411764705882,
      "eval_logits/chosen": -2.3386030197143555,
      "eval_logits/rejected": -1.7669346332550049,
      "eval_logps/chosen": -205.10494995117188,
      "eval_logps/rejected": -162.58966064453125,
      "eval_loss": 0.5985996127128601,
      "eval_rewards/accuracies": 0.5555555820465088,
      "eval_rewards/chosen": 0.07093840092420578,
      "eval_rewards/margins": 0.245022714138031,
      "eval_rewards/rejected": -0.17408432066440582,
      "eval_runtime": 10.3489,
      "eval_samples_per_second": 1.643,
      "eval_steps_per_second": 0.87,
      "step": 9
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}