{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9991836734693877,
  "eval_steps": 500,
  "global_step": 153,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 3.125e-08,
      "logits/chosen": -2.8377254009246826,
      "logits/rejected": -2.8079206943511963,
      "logps/chosen": -250.72598266601562,
      "logps/rejected": -220.86981201171875,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.1249999999999997e-07,
      "logits/chosen": -2.818211555480957,
      "logits/rejected": -2.7407193183898926,
      "logps/chosen": -223.67410278320312,
      "logps/rejected": -244.5283966064453,
      "loss": 0.6931,
      "rewards/accuracies": 0.4305555522441864,
      "rewards/chosen": 0.00010981507512042299,
      "rewards/margins": -0.0004382279294077307,
      "rewards/rejected": 0.000548043055459857,
      "step": 10
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.989490450759331e-07,
      "logits/chosen": -2.7361624240875244,
      "logits/rejected": -2.7406516075134277,
      "logps/chosen": -244.0074462890625,
      "logps/rejected": -265.5520935058594,
      "loss": 0.6925,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": 0.017707763239741325,
      "rewards/margins": 0.0019056980963796377,
      "rewards/rejected": 0.015802066773176193,
      "step": 20
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.872270441827174e-07,
      "logits/chosen": -2.6628098487854004,
      "logits/rejected": -2.6667449474334717,
      "logps/chosen": -242.32473754882812,
      "logps/rejected": -298.2237243652344,
      "loss": 0.6901,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": 0.06090053170919418,
      "rewards/margins": 0.005070054437965155,
      "rewards/rejected": 0.055830471217632294,
      "step": 30
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.6308512113530063e-07,
      "logits/chosen": -2.654808521270752,
      "logits/rejected": -2.6328892707824707,
      "logps/chosen": -261.2207336425781,
      "logps/rejected": -291.29718017578125,
      "loss": 0.6849,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": 0.03462223336100578,
      "rewards/margins": 0.017941609025001526,
      "rewards/rejected": 0.016680624336004257,
      "step": 40
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.277872161641681e-07,
      "logits/chosen": -2.644195079803467,
      "logits/rejected": -2.6148362159729004,
      "logps/chosen": -236.685546875,
      "logps/rejected": -253.20339965820312,
      "loss": 0.676,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.007446089293807745,
      "rewards/margins": 0.022974317893385887,
      "rewards/rejected": -0.03042040765285492,
      "step": 50
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.8318133624280046e-07,
      "logits/chosen": -2.6215896606445312,
      "logits/rejected": -2.611788749694824,
      "logps/chosen": -217.6464385986328,
      "logps/rejected": -262.92156982421875,
      "loss": 0.6697,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.04969613626599312,
      "rewards/margins": 0.06476574391126633,
      "rewards/rejected": -0.11446187645196915,
      "step": 60
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.316028034595861e-07,
      "logits/chosen": -2.5813162326812744,
      "logits/rejected": -2.550109386444092,
      "logps/chosen": -243.3549041748047,
      "logps/rejected": -266.4927978515625,
      "loss": 0.6645,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.09574780613183975,
      "rewards/margins": 0.07986501604318619,
      "rewards/rejected": -0.17561282217502594,
      "step": 70
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.7575199021178855e-07,
      "logits/chosen": -2.462153911590576,
      "logits/rejected": -2.461726665496826,
      "logps/chosen": -256.2433166503906,
      "logps/rejected": -277.47711181640625,
      "loss": 0.6609,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.27105945348739624,
      "rewards/margins": 0.08728176355361938,
      "rewards/rejected": -0.35834115743637085,
      "step": 80
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.1855294234408068e-07,
      "logits/chosen": -2.4760749340057373,
      "logits/rejected": -2.440697431564331,
      "logps/chosen": -245.5635986328125,
      "logps/rejected": -265.27850341796875,
      "loss": 0.6553,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -0.1536487340927124,
      "rewards/margins": 0.0763266533613205,
      "rewards/rejected": -0.2299753725528717,
      "step": 90
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.6300029195778453e-07,
      "logits/chosen": -2.3994648456573486,
      "logits/rejected": -2.3992934226989746,
      "logps/chosen": -253.02627563476562,
      "logps/rejected": -288.02374267578125,
      "loss": 0.6533,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.21652010083198547,
      "rewards/margins": 0.1159961000084877,
      "rewards/rejected": -0.3325161933898926,
      "step": 100
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.1200247470632392e-07,
      "logits/chosen": -2.3692357540130615,
      "logits/rejected": -2.3380370140075684,
      "logps/chosen": -245.78970336914062,
      "logps/rejected": -253.38925170898438,
      "loss": 0.6503,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.18969933688640594,
      "rewards/margins": 0.09776748716831207,
      "rewards/rejected": -0.287466824054718,
      "step": 110
    },
    {
      "epoch": 0.78,
      "learning_rate": 6.822945986946385e-08,
      "logits/chosen": -2.4934608936309814,
      "logits/rejected": -2.4202685356140137,
      "logps/chosen": -278.57794189453125,
      "logps/rejected": -307.510986328125,
      "loss": 0.6486,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.16311107575893402,
      "rewards/margins": 0.1353956162929535,
      "rewards/rejected": -0.2985066771507263,
      "step": 120
    },
    {
      "epoch": 0.85,
      "learning_rate": 3.397296523427806e-08,
      "logits/chosen": -2.335261821746826,
      "logits/rejected": -2.3717098236083984,
      "logps/chosen": -279.692626953125,
      "logps/rejected": -303.536865234375,
      "loss": 0.6376,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.18839788436889648,
      "rewards/margins": 0.11289455741643906,
      "rewards/rejected": -0.30129244923591614,
      "step": 130
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.1026475173977978e-08,
      "logits/chosen": -2.3872182369232178,
      "logits/rejected": -2.3598804473876953,
      "logps/chosen": -270.6761169433594,
      "logps/rejected": -299.95648193359375,
      "loss": 0.6474,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.2340303361415863,
      "rewards/margins": 0.1268264353275299,
      "rewards/rejected": -0.3608567714691162,
      "step": 140
    },
    {
      "epoch": 0.98,
      "learning_rate": 5.913435276374834e-10,
      "logits/chosen": -2.29586124420166,
      "logits/rejected": -2.324831485748291,
      "logps/chosen": -239.48080444335938,
      "logps/rejected": -296.5132751464844,
      "loss": 0.6436,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.19154903292655945,
      "rewards/margins": 0.1397678405046463,
      "rewards/rejected": -0.33131688833236694,
      "step": 150
    },
    {
      "epoch": 1.0,
      "step": 153,
      "total_flos": 0.0,
      "train_loss": 0.6648475749819887,
      "train_runtime": 9061.6732,
      "train_samples_per_second": 2.163,
      "train_steps_per_second": 0.017
    }
  ],
  "logging_steps": 10,
  "max_steps": 153,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}