{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.08820286659316427,
  "eval_steps": 500,
  "global_step": 60,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00735023888276369,
      "grad_norm": 4.643449306488037,
      "learning_rate": 4.999333020279093e-05,
      "logits/chosen": 21.724529266357422,
      "logits/rejected": 21.258750915527344,
      "logps/chosen": -288.99212646484375,
      "logps/rejected": -283.6695556640625,
      "loss": 0.5053,
      "num_input_tokens_seen": 56304,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": 2.1427507400512695,
      "rewards/margins": 2.0866787433624268,
      "rewards/rejected": 0.05607207491993904,
      "step": 5
    },
    {
      "epoch": 0.01470047776552738,
      "grad_norm": 0.004002333618700504,
      "learning_rate": 4.997332437005931e-05,
      "logits/chosen": 20.89365005493164,
      "logits/rejected": 20.077491760253906,
      "logps/chosen": -278.5107421875,
      "logps/rejected": -335.56353759765625,
      "loss": 0.0038,
      "num_input_tokens_seen": 111536,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.714799404144287,
      "rewards/margins": 10.309619903564453,
      "rewards/rejected": -7.594820499420166,
      "step": 10
    },
    {
      "epoch": 0.022050716648291068,
      "grad_norm": 0.003606519429013133,
      "learning_rate": 4.9939993176592926e-05,
      "logits/chosen": 21.55825424194336,
      "logits/rejected": 19.04710578918457,
      "logps/chosen": -306.679443359375,
      "logps/rejected": -402.765380859375,
      "loss": 0.0005,
      "num_input_tokens_seen": 170304,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.9726320505142212,
      "rewards/margins": 15.496417999267578,
      "rewards/rejected": -13.523785591125488,
      "step": 15
    },
    {
      "epoch": 0.02940095553105476,
      "grad_norm": 0.00013470888370648026,
      "learning_rate": 4.989335440737586e-05,
      "logits/chosen": 22.240352630615234,
      "logits/rejected": 20.095966339111328,
      "logps/chosen": -334.5459289550781,
      "logps/rejected": -465.93316650390625,
      "loss": 0.0,
      "num_input_tokens_seen": 231248,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.5124895572662354,
      "rewards/margins": 18.956003189086914,
      "rewards/rejected": -17.443511962890625,
      "step": 20
    },
    {
      "epoch": 0.036751194413818446,
      "grad_norm": 1.964755256267381e-06,
      "learning_rate": 4.9833432948098754e-05,
      "logits/chosen": 21.701492309570312,
      "logits/rejected": 19.725160598754883,
      "logps/chosen": -296.9284362792969,
      "logps/rejected": -462.3471984863281,
      "loss": 0.0004,
      "num_input_tokens_seen": 286288,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.434482216835022,
      "rewards/margins": 21.793750762939453,
      "rewards/rejected": -20.359268188476562,
      "step": 25
    },
    {
      "epoch": 0.044101433296582136,
      "grad_norm": 1.0419856153021101e-05,
      "learning_rate": 4.976026077188013e-05,
      "logits/chosen": 22.023757934570312,
      "logits/rejected": 19.70968246459961,
      "logps/chosen": -338.1859130859375,
      "logps/rejected": -502.81890869140625,
      "loss": 0.0,
      "num_input_tokens_seen": 347904,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.06728994846344,
      "rewards/margins": 21.31583023071289,
      "rewards/rejected": -20.248538970947266,
      "step": 30
    },
    {
      "epoch": 0.05145167217934583,
      "grad_norm": 0.00010334135731682181,
      "learning_rate": 4.967387692220614e-05,
      "logits/chosen": 21.776086807250977,
      "logits/rejected": 21.1005802154541,
      "logps/chosen": -308.8163757324219,
      "logps/rejected": -506.4019775390625,
      "loss": 0.0,
      "num_input_tokens_seen": 404624,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.37391197681427,
      "rewards/margins": 23.01706314086914,
      "rewards/rejected": -21.643150329589844,
      "step": 35
    },
    {
      "epoch": 0.05880191106210952,
      "grad_norm": 1.0441175390951685e-06,
      "learning_rate": 4.957432749209755e-05,
      "logits/chosen": 23.046411514282227,
      "logits/rejected": 18.835281372070312,
      "logps/chosen": -345.61572265625,
      "logps/rejected": -501.54962158203125,
      "loss": 0.0,
      "num_input_tokens_seen": 467984,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.5436484217643738,
      "rewards/margins": 22.15255355834961,
      "rewards/rejected": -21.608903884887695,
      "step": 40
    },
    {
      "epoch": 0.06615214994487321,
      "grad_norm": 4.266040923539549e-05,
      "learning_rate": 4.946166559951523e-05,
      "logits/chosen": 21.52145767211914,
      "logits/rejected": 21.178810119628906,
      "logps/chosen": -311.7443542480469,
      "logps/rejected": -516.5821533203125,
      "loss": 0.0,
      "num_input_tokens_seen": 526448,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.872286856174469,
      "rewards/margins": 22.805173873901367,
      "rewards/rejected": -21.93288803100586,
      "step": 45
    },
    {
      "epoch": 0.07350238882763689,
      "grad_norm": 1.2180415751572582e-06,
      "learning_rate": 4.933595135901732e-05,
      "logits/chosen": 21.246919631958008,
      "logits/rejected": 21.005220413208008,
      "logps/chosen": -306.83197021484375,
      "logps/rejected": -517.5718994140625,
      "loss": 0.0,
      "num_input_tokens_seen": 583680,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.57448410987854,
      "rewards/margins": 23.323020935058594,
      "rewards/rejected": -22.748537063598633,
      "step": 50
    },
    {
      "epoch": 0.08085262771040058,
      "grad_norm": 7.444796210620552e-07,
      "learning_rate": 4.9197251849683066e-05,
      "logits/chosen": 21.65299415588379,
      "logits/rejected": 19.46175765991211,
      "logps/chosen": -299.33660888671875,
      "logps/rejected": -490.9933166503906,
      "loss": 0.0,
      "num_input_tokens_seen": 640432,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.3190252780914307,
      "rewards/margins": 24.110233306884766,
      "rewards/rejected": -22.791208267211914,
      "step": 55
    },
    {
      "epoch": 0.08820286659316427,
      "grad_norm": 1.690499288997671e-06,
      "learning_rate": 4.9045641079320484e-05,
      "logits/chosen": 22.51614761352539,
      "logits/rejected": 21.07533836364746,
      "logps/chosen": -339.6164855957031,
      "logps/rejected": -536.0887451171875,
      "loss": 0.0,
      "num_input_tokens_seen": 701920,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.7054795026779175,
      "rewards/margins": 23.903228759765625,
      "rewards/rejected": -23.19774627685547,
      "step": 60
    }
  ],
  "logging_steps": 5,
  "max_steps": 680,
  "num_input_tokens_seen": 701920,
  "num_train_epochs": 1,
  "save_steps": 30,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.1882328485658624e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}