{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.984,
  "eval_steps": 100,
  "global_step": 124,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 3.846153846153847e-07,
      "logits/chosen": 0.4176342189311981,
      "logits/rejected": 0.3709704279899597,
      "logps/chosen": -211.92245483398438,
      "logps/rejected": -192.94674682617188,
      "loss": 0.0102,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.16,
      "learning_rate": 3.846153846153847e-06,
      "logits/chosen": 0.11221448332071304,
      "logits/rejected": 0.2316199392080307,
      "logps/chosen": -163.23516845703125,
      "logps/rejected": -129.18174743652344,
      "loss": 0.0101,
      "rewards/accuracies": 0.4097222089767456,
      "rewards/chosen": 0.002231667749583721,
      "rewards/margins": 0.0012476635165512562,
      "rewards/rejected": 0.000984004233032465,
      "step": 10
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.951096619903317e-06,
      "logits/chosen": 0.11361704021692276,
      "logits/rejected": 0.05081893131136894,
      "logps/chosen": -177.82867431640625,
      "logps/rejected": -143.34634399414062,
      "loss": 0.0103,
      "rewards/accuracies": 0.40625,
      "rewards/chosen": -0.0008938458049669862,
      "rewards/margins": 0.00015756110951770097,
      "rewards/rejected": -0.0010514067253097892,
      "step": 20
    },
    {
      "epoch": 0.48,
      "learning_rate": 4.716164218065246e-06,
      "logits/chosen": 0.1756063550710678,
      "logits/rejected": 0.23870661854743958,
      "logps/chosen": -182.95594787597656,
      "logps/rejected": -149.51669311523438,
      "loss": 0.0104,
      "rewards/accuracies": 0.38749998807907104,
      "rewards/chosen": 0.0036168531514704227,
      "rewards/margins": 8.136676478898153e-05,
      "rewards/rejected": 0.0035354860592633486,
      "step": 30
    },
    {
      "epoch": 0.64,
      "learning_rate": 4.3048902348863116e-06,
      "logits/chosen": 0.23890891671180725,
      "logits/rejected": 0.15510627627372742,
      "logps/chosen": -198.42312622070312,
      "logps/rejected": -165.66856384277344,
      "loss": 0.0103,
      "rewards/accuracies": 0.41874998807907104,
      "rewards/chosen": -0.000657144351862371,
      "rewards/margins": 0.0006801211275160313,
      "rewards/rejected": -0.0013372651301324368,
      "step": 40
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.7500000000000005e-06,
      "logits/chosen": 0.0918453186750412,
      "logits/rejected": 0.22310352325439453,
      "logps/chosen": -167.31259155273438,
      "logps/rejected": -152.1429443359375,
      "loss": 0.0103,
      "rewards/accuracies": 0.4000000059604645,
      "rewards/chosen": -0.0004377079603727907,
      "rewards/margins": 0.00018744784756563604,
      "rewards/rejected": -0.0006251559825614095,
      "step": 50
    },
    {
      "epoch": 0.96,
      "learning_rate": 3.0956464785579125e-06,
      "logits/chosen": 0.17539598047733307,
      "logits/rejected": 0.15246877074241638,
      "logps/chosen": -172.65084838867188,
      "logps/rejected": -149.18576049804688,
      "loss": 0.0102,
      "rewards/accuracies": 0.48124998807907104,
      "rewards/chosen": 0.00030969156068749726,
      "rewards/margins": 0.0009550409158691764,
      "rewards/rejected": -0.0006453495589084923,
      "step": 60
    },
    {
      "epoch": 1.12,
      "learning_rate": 2.39389699200963e-06,
      "logits/chosen": 0.2547686994075775,
      "logits/rejected": 0.22828814387321472,
      "logps/chosen": -180.34805297851562,
      "logps/rejected": -161.528076171875,
      "loss": 0.0105,
      "rewards/accuracies": 0.375,
      "rewards/chosen": -0.0031430914532393217,
      "rewards/margins": -0.0006838366389274597,
      "rewards/rejected": -0.002459254814311862,
      "step": 70
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.700590188571887e-06,
      "logits/chosen": 0.24805791676044464,
      "logits/rejected": 0.14473767578601837,
      "logps/chosen": -168.27664184570312,
      "logps/rejected": -145.95330810546875,
      "loss": 0.0107,
      "rewards/accuracies": 0.38749998807907104,
      "rewards/chosen": -0.001294322544708848,
      "rewards/margins": -0.0014985360903665423,
      "rewards/rejected": 0.00020421366207301617,
      "step": 80
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0708929268538034e-06,
      "logits/chosen": 0.21019116044044495,
      "logits/rejected": 0.0877661406993866,
      "logps/chosen": -181.98239135742188,
      "logps/rejected": -144.48709106445312,
      "loss": 0.0103,
      "rewards/accuracies": 0.375,
      "rewards/chosen": -0.0010133858304470778,
      "rewards/margins": 0.00020684293122030795,
      "rewards/rejected": -0.0012202286161482334,
      "step": 90
    },
    {
      "epoch": 1.6,
      "learning_rate": 5.549106142039018e-07,
      "logits/chosen": 0.17486083507537842,
      "logits/rejected": 0.10187114775180817,
      "logps/chosen": -171.99929809570312,
      "logps/rejected": -156.14585876464844,
      "loss": 0.0104,
      "rewards/accuracies": 0.3812499940395355,
      "rewards/chosen": -0.00023761784541420639,
      "rewards/margins": -0.00010095858306158334,
      "rewards/rejected": -0.00013665916048921645,
      "step": 100
    },
    {
      "epoch": 1.6,
      "eval_logits/chosen": -0.010001667775213718,
      "eval_logits/rejected": 0.08886812627315521,
      "eval_logps/chosen": -306.4288635253906,
      "eval_logps/rejected": -278.7423095703125,
      "eval_loss": 0.01069465558975935,
      "eval_rewards/accuracies": 0.5145000219345093,
      "eval_rewards/chosen": -0.00046658667270094156,
      "eval_rewards/margins": 0.00048252198030240834,
      "eval_rewards/rejected": -0.0009491087403148413,
      "eval_runtime": 710.6081,
      "eval_samples_per_second": 2.814,
      "eval_steps_per_second": 0.704,
      "step": 100
    },
    {
      "epoch": 1.76,
      "learning_rate": 1.937002879188285e-07,
      "logits/chosen": 0.17259174585342407,
      "logits/rejected": 0.22864675521850586,
      "logps/chosen": -194.3363494873047,
      "logps/rejected": -159.630126953125,
      "loss": 0.0099,
      "rewards/accuracies": 0.4375,
      "rewards/chosen": -0.00021430714696180075,
      "rewards/margins": 0.0019922677893191576,
      "rewards/rejected": -0.002206574659794569,
      "step": 110
    },
    {
      "epoch": 1.92,
      "learning_rate": 1.6003680950742728e-08,
      "logits/chosen": 0.14167846739292145,
      "logits/rejected": 0.09132746607065201,
      "logps/chosen": -186.17977905273438,
      "logps/rejected": -150.71856689453125,
      "loss": 0.0102,
      "rewards/accuracies": 0.46875,
      "rewards/chosen": -0.0018472172087058425,
      "rewards/margins": 0.0006628064438700676,
      "rewards/rejected": -0.002510023768991232,
      "step": 120
    },
    {
      "epoch": 1.98,
      "step": 124,
      "total_flos": 0.0,
      "train_loss": 0.010306676939850854,
      "train_runtime": 2255.3236,
      "train_samples_per_second": 0.887,
      "train_steps_per_second": 0.055
    }
  ],
  "logging_steps": 10,
  "max_steps": 124,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}