{
  "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": 66.08671668865084,
      "learning_rate": 4.545454545454545e-08,
      "logits/chosen": 117.53560638427734,
      "logits/rejected": 126.8960952758789,
      "logps/chosen": -335.40118408203125,
      "logps/rejected": -439.16552734375,
      "loss": 0.5,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.1895734597156398,
      "grad_norm": 69.395273460518,
      "learning_rate": 4.545454545454545e-07,
      "logits/chosen": 135.0714569091797,
      "logits/rejected": 138.4192657470703,
      "logps/chosen": -394.4978332519531,
      "logps/rejected": -438.8009338378906,
      "loss": 0.4962,
      "rewards/accuracies": 0.4652777910232544,
      "rewards/chosen": 0.07493551820516586,
      "rewards/margins": 0.048605356365442276,
      "rewards/rejected": 0.026330159977078438,
      "step": 10
    },
    {
      "epoch": 0.3791469194312796,
      "grad_norm": 29.977088974021598,
      "learning_rate": 4.885348141000122e-07,
      "logits/chosen": 122.71434020996094,
      "logits/rejected": 126.32965087890625,
      "logps/chosen": -353.5599060058594,
      "logps/rejected": -406.46490478515625,
      "loss": 0.411,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": 0.9426037073135376,
      "rewards/margins": 0.2662777602672577,
      "rewards/rejected": 0.676325798034668,
      "step": 20
    },
    {
      "epoch": 0.5687203791469194,
      "grad_norm": 25.169253527598716,
      "learning_rate": 4.5025027361734613e-07,
      "logits/chosen": 144.8006134033203,
      "logits/rejected": 138.1434326171875,
      "logps/chosen": -380.2307434082031,
      "logps/rejected": -436.6331481933594,
      "loss": 0.3519,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": 0.6117245554924011,
      "rewards/margins": 1.4432713985443115,
      "rewards/rejected": -0.8315467834472656,
      "step": 30
    },
    {
      "epoch": 0.7582938388625592,
      "grad_norm": 24.67006837987384,
      "learning_rate": 3.893311157806091e-07,
      "logits/chosen": 136.48574829101562,
      "logits/rejected": 125.43827819824219,
      "logps/chosen": -322.3842468261719,
      "logps/rejected": -364.0414733886719,
      "loss": 0.3425,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": 1.5904960632324219,
      "rewards/margins": 1.8988056182861328,
      "rewards/rejected": -0.3083093464374542,
      "step": 40
    },
    {
      "epoch": 0.9478672985781991,
      "grad_norm": 25.00841040201808,
      "learning_rate": 3.126631330646801e-07,
      "logits/chosen": 154.9099884033203,
      "logits/rejected": 158.31307983398438,
      "logps/chosen": -383.2764892578125,
      "logps/rejected": -484.3291015625,
      "loss": 0.3108,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 1.793367624282837,
      "rewards/margins": 2.193502902984619,
      "rewards/rejected": -0.4001353681087494,
      "step": 50
    },
    {
      "epoch": 1.1374407582938388,
      "grad_norm": 20.059631761558116,
      "learning_rate": 2.2891223348923882e-07,
      "logits/chosen": 145.4669647216797,
      "logits/rejected": 149.66366577148438,
      "logps/chosen": -360.8850402832031,
      "logps/rejected": -458.97705078125,
      "loss": 0.2596,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": 1.8560327291488647,
      "rewards/margins": 2.9866089820861816,
      "rewards/rejected": -1.1305763721466064,
      "step": 60
    },
    {
      "epoch": 1.3270142180094786,
      "grad_norm": 19.000102691830318,
      "learning_rate": 1.4754491880085317e-07,
      "logits/chosen": 140.24868774414062,
      "logits/rejected": 141.07412719726562,
      "logps/chosen": -329.3720703125,
      "logps/rejected": -430.41650390625,
      "loss": 0.2358,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": 1.8885892629623413,
      "rewards/margins": 2.7337210178375244,
      "rewards/rejected": -0.8451315760612488,
      "step": 70
    },
    {
      "epoch": 1.5165876777251186,
      "grad_norm": 20.284917774527994,
      "learning_rate": 7.775827023107834e-08,
      "logits/chosen": 127.765625,
      "logits/rejected": 143.33151245117188,
      "logps/chosen": -311.8875427246094,
      "logps/rejected": -435.666015625,
      "loss": 0.2087,
      "rewards/accuracies": 0.8500000238418579,
      "rewards/chosen": 2.0881741046905518,
      "rewards/margins": 3.2095909118652344,
      "rewards/rejected": -1.121416687965393,
      "step": 80
    },
    {
      "epoch": 1.7061611374407581,
      "grad_norm": 20.77837485504225,
      "learning_rate": 2.7440387297912122e-08,
      "logits/chosen": 128.07351684570312,
      "logits/rejected": 140.38681030273438,
      "logps/chosen": -341.97064208984375,
      "logps/rejected": -461.2508850097656,
      "loss": 0.2054,
      "rewards/accuracies": 0.84375,
      "rewards/chosen": 2.22301983833313,
      "rewards/margins": 3.40147066116333,
      "rewards/rejected": -1.1784509420394897,
      "step": 90
    },
    {
      "epoch": 1.8957345971563981,
      "grad_norm": 19.748881150000962,
      "learning_rate": 2.27878296044029e-09,
      "logits/chosen": 134.47386169433594,
      "logits/rejected": 134.50311279296875,
      "logps/chosen": -334.57818603515625,
      "logps/rejected": -429.48236083984375,
      "loss": 0.1946,
      "rewards/accuracies": 0.84375,
      "rewards/chosen": 2.3698461055755615,
      "rewards/margins": 3.005765438079834,
      "rewards/rejected": -0.6359192728996277,
      "step": 100
    },
    {
      "epoch": 1.8957345971563981,
      "eval_logits/chosen": 113.83429718017578,
      "eval_logits/rejected": 108.64144134521484,
      "eval_logps/chosen": -333.18048095703125,
      "eval_logps/rejected": -365.2297668457031,
      "eval_loss": 0.2941707372665405,
      "eval_rewards/accuracies": 0.6458333134651184,
      "eval_rewards/chosen": 1.5243864059448242,
      "eval_rewards/margins": 1.6823266744613647,
      "eval_rewards/rejected": -0.15794026851654053,
      "eval_runtime": 116.4001,
      "eval_samples_per_second": 6.443,
      "eval_steps_per_second": 0.206,
      "step": 100
    },
    {
      "epoch": 1.971563981042654,
      "step": 104,
      "total_flos": 0.0,
      "train_loss": 0.2986852119748409,
      "train_runtime": 2293.3212,
      "train_samples_per_second": 5.887,
      "train_steps_per_second": 0.045
    }
  ],
  "logging_steps": 10,
  "max_steps": 104,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}