{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 1000,
  "global_step": 125,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "grad_norm": 3.554304809363554,
      "learning_rate": 3.846153846153847e-07,
      "logits/chosen": -1.7501684427261353,
      "logits/rejected": -1.5946102142333984,
      "logps/chosen": -84.6928939819336,
      "logps/rejected": -66.21642303466797,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.4,
      "grad_norm": 4.900524905972191,
      "learning_rate": 3.846153846153847e-06,
      "logits/chosen": -1.7080682516098022,
      "logits/rejected": -1.6930590867996216,
      "logps/chosen": -71.15042114257812,
      "logps/rejected": -81.29680633544922,
      "loss": 0.6627,
      "rewards/accuracies": 0.8055555820465088,
      "rewards/chosen": 0.040672361850738525,
      "rewards/margins": 0.06852560490369797,
      "rewards/rejected": -0.027853243052959442,
      "step": 10
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.17197492369867923,
      "learning_rate": 4.9519632010080765e-06,
      "logits/chosen": -1.501769781112671,
      "logits/rejected": -1.5045287609100342,
      "logps/chosen": -55.158836364746094,
      "logps/rejected": -317.1681823730469,
      "loss": 0.2119,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.2232411801815033,
      "rewards/margins": 2.6507387161254883,
      "rewards/rejected": -2.427497386932373,
      "step": 20
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.014772261199203527,
      "learning_rate": 4.721114089947181e-06,
      "logits/chosen": -0.9690781831741333,
      "logits/rejected": -1.0795819759368896,
      "logps/chosen": -249.5034637451172,
      "logps/rejected": -1324.19091796875,
      "loss": 0.0006,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.7644612789154053,
      "rewards/margins": 10.673975944519043,
      "rewards/rejected": -12.438436508178711,
      "step": 30
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.0033918924347121214,
      "learning_rate": 4.316650805085068e-06,
      "logits/chosen": -1.2455154657363892,
      "logits/rejected": -1.294455647468567,
      "logps/chosen": -155.76324462890625,
      "logps/rejected": -1702.1519775390625,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.8099241256713867,
      "rewards/margins": 15.425918579101562,
      "rewards/rejected": -16.235843658447266,
      "step": 40
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.002248346138217333,
      "learning_rate": 3.770188363116324e-06,
      "logits/chosen": -1.3940218687057495,
      "logits/rejected": -1.3445849418640137,
      "logps/chosen": -162.5945587158203,
      "logps/rejected": -1922.8701171875,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.868628978729248,
      "rewards/margins": 17.592626571655273,
      "rewards/rejected": -18.46125602722168,
      "step": 50
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.001498031862466307,
      "learning_rate": 3.1244411954180677e-06,
      "logits/chosen": -1.5801560878753662,
      "logits/rejected": -1.383643388748169,
      "logps/chosen": -182.91412353515625,
      "logps/rejected": -2047.5972900390625,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.0077913999557495,
      "rewards/margins": 18.757488250732422,
      "rewards/rejected": -19.765277862548828,
      "step": 60
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.0017507720015667862,
      "learning_rate": 2.429884359310328e-06,
      "logits/chosen": -1.3874660730361938,
      "logits/rejected": -1.264624834060669,
      "logps/chosen": -170.91595458984375,
      "logps/rejected": -1971.678466796875,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.9257668256759644,
      "rewards/margins": 18.046239852905273,
      "rewards/rejected": -18.972007751464844,
      "step": 70
    },
    {
      "epoch": 3.2,
      "grad_norm": 0.0012759873530227032,
      "learning_rate": 1.7408081372259633e-06,
      "logits/chosen": -1.3026721477508545,
      "logits/rejected": -1.211972951889038,
      "logps/chosen": -161.51548767089844,
      "logps/rejected": -1921.3414306640625,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.8770667314529419,
      "rewards/margins": 17.575910568237305,
      "rewards/rejected": -18.452978134155273,
      "step": 80
    },
    {
      "epoch": 3.6,
      "grad_norm": 0.0015468265273843927,
      "learning_rate": 1.1110744174509952e-06,
      "logits/chosen": -1.4862353801727295,
      "logits/rejected": -1.386122226715088,
      "logps/chosen": -173.3203887939453,
      "logps/rejected": -2036.2054443359375,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.9565305709838867,
      "rewards/margins": 18.671863555908203,
      "rewards/rejected": -19.628393173217773,
      "step": 90
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.0013608753936275465,
      "learning_rate": 5.899065604459814e-07,
      "logits/chosen": -1.2386879920959473,
      "logits/rejected": -1.2031110525131226,
      "logps/chosen": -157.1131591796875,
      "logps/rejected": -1883.630126953125,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.8118581771850586,
      "rewards/margins": 17.229957580566406,
      "rewards/rejected": -18.041812896728516,
      "step": 100
    },
    {
      "epoch": 4.4,
      "grad_norm": 0.000781528269573621,
      "learning_rate": 2.1804183734670277e-07,
      "logits/chosen": -1.2313117980957031,
      "logits/rejected": -1.2176135778427124,
      "logps/chosen": -154.5117645263672,
      "logps/rejected": -1841.933837890625,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.8211410641670227,
      "rewards/margins": 16.808500289916992,
      "rewards/rejected": -17.629640579223633,
      "step": 110
    },
    {
      "epoch": 4.8,
      "grad_norm": 0.0015269909111218947,
      "learning_rate": 2.454718665888589e-08,
      "logits/chosen": -1.4756698608398438,
      "logits/rejected": -1.2926015853881836,
      "logps/chosen": -181.42701721191406,
      "logps/rejected": -1979.6390380859375,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.0040886402130127,
      "rewards/margins": 18.079927444458008,
      "rewards/rejected": -19.084014892578125,
      "step": 120
    },
    {
      "epoch": 5.0,
      "step": 125,
      "total_flos": 0.0,
      "train_loss": 0.07030335423257202,
      "train_runtime": 8011.3193,
      "train_samples_per_second": 0.499,
      "train_steps_per_second": 0.016
    }
  ],
  "logging_steps": 10,
  "max_steps": 125,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}