{
  "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": 66.10108494411732,
      "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": 63.35028062693078,
      "learning_rate": 4.545454545454545e-07,
      "logits/chosen": 135.05430603027344,
      "logits/rejected": 138.40155029296875,
      "logps/chosen": -395.61114501953125,
      "logps/rejected": -438.9917907714844,
      "loss": 0.4945,
      "rewards/accuracies": 0.4097222089767456,
      "rewards/chosen": 0.019270293414592743,
      "rewards/margins": 0.0024831907358020544,
      "rewards/rejected": 0.016787106171250343,
      "step": 10
    },
    {
      "epoch": 0.3791469194312796,
      "grad_norm": 30.009186267506557,
      "learning_rate": 4.885348141000122e-07,
      "logits/chosen": 122.25505065917969,
      "logits/rejected": 125.93106842041016,
      "logps/chosen": -353.450439453125,
      "logps/rejected": -405.93060302734375,
      "loss": 0.41,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": 0.9480773210525513,
      "rewards/margins": 0.24503827095031738,
      "rewards/rejected": 0.7030390501022339,
      "step": 20
    },
    {
      "epoch": 0.5687203791469194,
      "grad_norm": 25.296178390352267,
      "learning_rate": 4.5025027361734613e-07,
      "logits/chosen": 143.24757385253906,
      "logits/rejected": 136.6645965576172,
      "logps/chosen": -379.17156982421875,
      "logps/rejected": -435.64471435546875,
      "loss": 0.3504,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": 0.6646836996078491,
      "rewards/margins": 1.4468095302581787,
      "rewards/rejected": -0.7821259498596191,
      "step": 30
    },
    {
      "epoch": 0.7582938388625592,
      "grad_norm": 24.801763821402933,
      "learning_rate": 3.893311157806091e-07,
      "logits/chosen": 134.1096649169922,
      "logits/rejected": 123.01851654052734,
      "logps/chosen": -322.86090087890625,
      "logps/rejected": -364.0258483886719,
      "loss": 0.3439,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": 1.5666635036468506,
      "rewards/margins": 1.874193549156189,
      "rewards/rejected": -0.30753010511398315,
      "step": 40
    },
    {
      "epoch": 0.9478672985781991,
      "grad_norm": 25.32318087440862,
      "learning_rate": 3.126631330646801e-07,
      "logits/chosen": 153.5646514892578,
      "logits/rejected": 156.97068786621094,
      "logps/chosen": -381.78765869140625,
      "logps/rejected": -481.5089416503906,
      "loss": 0.3097,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": 1.8678098917007446,
      "rewards/margins": 2.126939296722412,
      "rewards/rejected": -0.2591293752193451,
      "step": 50
    },
    {
      "epoch": 1.1374407582938388,
      "grad_norm": 20.749611901077518,
      "learning_rate": 2.2891223348923882e-07,
      "logits/chosen": 143.95127868652344,
      "logits/rejected": 148.13980102539062,
      "logps/chosen": -360.61602783203125,
      "logps/rejected": -458.1600036621094,
      "loss": 0.2569,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 1.8694829940795898,
      "rewards/margins": 2.9592061042785645,
      "rewards/rejected": -1.0897233486175537,
      "step": 60
    },
    {
      "epoch": 1.3270142180094786,
      "grad_norm": 18.9652516165648,
      "learning_rate": 1.4754491880085317e-07,
      "logits/chosen": 138.2149658203125,
      "logits/rejected": 139.26931762695312,
      "logps/chosen": -327.8791198730469,
      "logps/rejected": -428.8116760253906,
      "loss": 0.2361,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": 1.9632362127304077,
      "rewards/margins": 2.72812819480896,
      "rewards/rejected": -0.7648921608924866,
      "step": 70
    },
    {
      "epoch": 1.5165876777251186,
      "grad_norm": 19.639368765017288,
      "learning_rate": 7.775827023107834e-08,
      "logits/chosen": 125.30033111572266,
      "logits/rejected": 141.14437866210938,
      "logps/chosen": -311.32623291015625,
      "logps/rejected": -434.94378662109375,
      "loss": 0.2086,
      "rewards/accuracies": 0.856249988079071,
      "rewards/chosen": 2.116239547729492,
      "rewards/margins": 3.2015445232391357,
      "rewards/rejected": -1.0853049755096436,
      "step": 80
    },
    {
      "epoch": 1.7061611374407581,
      "grad_norm": 21.053763020863467,
      "learning_rate": 2.7440387297912122e-08,
      "logits/chosen": 125.8764419555664,
      "logits/rejected": 138.4185028076172,
      "logps/chosen": -340.38427734375,
      "logps/rejected": -458.72796630859375,
      "loss": 0.2018,
      "rewards/accuracies": 0.8500000238418579,
      "rewards/chosen": 2.3023386001586914,
      "rewards/margins": 3.3546459674835205,
      "rewards/rejected": -1.052307367324829,
      "step": 90
    },
    {
      "epoch": 1.8957345971563981,
      "grad_norm": 20.135689425911313,
      "learning_rate": 2.27878296044029e-09,
      "logits/chosen": 132.50669860839844,
      "logits/rejected": 132.55801391601562,
      "logps/chosen": -332.8918151855469,
      "logps/rejected": -427.8233947753906,
      "loss": 0.1942,
      "rewards/accuracies": 0.831250011920929,
      "rewards/chosen": 2.4541642665863037,
      "rewards/margins": 3.0071334838867188,
      "rewards/rejected": -0.5529693365097046,
      "step": 100
    },
    {
      "epoch": 1.8957345971563981,
      "eval_logits/chosen": 111.29891967773438,
      "eval_logits/rejected": 106.04138946533203,
      "eval_logps/chosen": -332.04876708984375,
      "eval_logps/rejected": -363.3305358886719,
      "eval_loss": 0.2925301492214203,
      "eval_rewards/accuracies": 0.6770833134651184,
      "eval_rewards/chosen": 1.5809730291366577,
      "eval_rewards/margins": 1.64395010471344,
      "eval_rewards/rejected": -0.06297732144594193,
      "eval_runtime": 121.1488,
      "eval_samples_per_second": 6.191,
      "eval_steps_per_second": 0.198,
      "step": 100
    },
    {
      "epoch": 1.971563981042654,
      "step": 104,
      "total_flos": 0.0,
      "train_loss": 0.29734468632019484,
      "train_runtime": 2303.4906,
      "train_samples_per_second": 5.861,
      "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
}