{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984301412872841,
  "eval_steps": 100,
  "global_step": 159,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 3.125e-07,
      "logits/chosen": -2.872462511062622,
      "logits/rejected": -2.8200550079345703,
      "logps/chosen": -279.5750732421875,
      "logps/rejected": -283.5914306640625,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.125e-06,
      "logits/chosen": -2.7532310485839844,
      "logits/rejected": -2.7174949645996094,
      "logps/chosen": -283.4659423828125,
      "logps/rejected": -245.05682373046875,
      "loss": 0.6929,
      "rewards/accuracies": 0.4652777910232544,
      "rewards/chosen": -0.00040846053161658347,
      "rewards/margins": 0.0003327548620291054,
      "rewards/rejected": -0.0007412154227495193,
      "step": 10
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.9903533134293035e-06,
      "logits/chosen": -2.7988767623901367,
      "logits/rejected": -2.745978832244873,
      "logps/chosen": -275.3612365722656,
      "logps/rejected": -259.2863464355469,
      "loss": 0.6891,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": 0.0042456588707864285,
      "rewards/margins": 0.009360184893012047,
      "rewards/rejected": -0.005114525556564331,
      "step": 20
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.882681251368549e-06,
      "logits/chosen": -2.750326633453369,
      "logits/rejected": -2.6968815326690674,
      "logps/chosen": -287.40570068359375,
      "logps/rejected": -264.3677978515625,
      "loss": 0.6796,
      "rewards/accuracies": 0.653124988079071,
      "rewards/chosen": 0.019095653668045998,
      "rewards/margins": 0.02568042278289795,
      "rewards/rejected": -0.006584773771464825,
      "step": 30
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.660472094042121e-06,
      "logits/chosen": -2.753675937652588,
      "logits/rejected": -2.7094359397888184,
      "logps/chosen": -269.2148742675781,
      "logps/rejected": -266.2108459472656,
      "loss": 0.6689,
      "rewards/accuracies": 0.625,
      "rewards/chosen": 0.008650553412735462,
      "rewards/margins": 0.03987262770533562,
      "rewards/rejected": -0.031222078949213028,
      "step": 40
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.33440758555951e-06,
      "logits/chosen": -2.700374126434326,
      "logits/rejected": -2.640312910079956,
      "logps/chosen": -279.00579833984375,
      "logps/rejected": -280.64959716796875,
      "loss": 0.657,
      "rewards/accuracies": 0.684374988079071,
      "rewards/chosen": -0.0027772835455834866,
      "rewards/margins": 0.06927188485860825,
      "rewards/rejected": -0.07204917818307877,
      "step": 50
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.92016186682789e-06,
      "logits/chosen": -2.695119857788086,
      "logits/rejected": -2.624582529067993,
      "logps/chosen": -274.4857482910156,
      "logps/rejected": -262.78082275390625,
      "loss": 0.6506,
      "rewards/accuracies": 0.7093750238418579,
      "rewards/chosen": 0.0005743972724303603,
      "rewards/margins": 0.09640298783779144,
      "rewards/rejected": -0.095828577876091,
      "step": 60
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.437648009023905e-06,
      "logits/chosen": -2.645156145095825,
      "logits/rejected": -2.5854098796844482,
      "logps/chosen": -281.52117919921875,
      "logps/rejected": -287.34521484375,
      "loss": 0.6423,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.0025214666966348886,
      "rewards/margins": 0.1480122059583664,
      "rewards/rejected": -0.15053364634513855,
      "step": 70
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.9100607788275547e-06,
      "logits/chosen": -2.600132703781128,
      "logits/rejected": -2.5598268508911133,
      "logps/chosen": -278.39520263671875,
      "logps/rejected": -274.490478515625,
      "loss": 0.6365,
      "rewards/accuracies": 0.684374988079071,
      "rewards/chosen": -0.08650828897953033,
      "rewards/margins": 0.12828321754932404,
      "rewards/rejected": -0.21479150652885437,
      "step": 80
    },
    {
      "epoch": 0.57,
      "learning_rate": 2.3627616503391813e-06,
      "logits/chosen": -2.6346802711486816,
      "logits/rejected": -2.561859130859375,
      "logps/chosen": -267.5691223144531,
      "logps/rejected": -296.9002380371094,
      "loss": 0.625,
      "rewards/accuracies": 0.684374988079071,
      "rewards/chosen": -0.10211894661188126,
      "rewards/margins": 0.14463460445404053,
      "rewards/rejected": -0.2467535436153412,
      "step": 90
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.8220596619089576e-06,
      "logits/chosen": -2.628725528717041,
      "logits/rejected": -2.5589451789855957,
      "logps/chosen": -290.6697082519531,
      "logps/rejected": -277.4861755371094,
      "loss": 0.6186,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -0.08957623690366745,
      "rewards/margins": 0.20158691704273224,
      "rewards/rejected": -0.2911631464958191,
      "step": 100
    },
    {
      "epoch": 0.63,
      "eval_logits/chosen": -2.6340699195861816,
      "eval_logits/rejected": -2.550675868988037,
      "eval_logps/chosen": -293.948974609375,
      "eval_logps/rejected": -284.51849365234375,
      "eval_loss": 0.627379298210144,
      "eval_rewards/accuracies": 0.6800000071525574,
      "eval_rewards/chosen": -0.09909208118915558,
      "eval_rewards/margins": 0.17093417048454285,
      "eval_rewards/rejected": -0.27002623677253723,
      "eval_runtime": 384.4479,
      "eval_samples_per_second": 5.202,
      "eval_steps_per_second": 0.65,
      "step": 100
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.3139467229135999e-06,
      "logits/chosen": -2.6164746284484863,
      "logits/rejected": -2.556546688079834,
      "logps/chosen": -296.1142272949219,
      "logps/rejected": -295.5721740722656,
      "loss": 0.6274,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.0782129243016243,
      "rewards/margins": 0.19299447536468506,
      "rewards/rejected": -0.27120739221572876,
      "step": 110
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.628481651367876e-07,
      "logits/chosen": -2.619702100753784,
      "logits/rejected": -2.5185346603393555,
      "logps/chosen": -305.4084167480469,
      "logps/rejected": -269.8453063964844,
      "loss": 0.6209,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.11033356189727783,
      "rewards/margins": 0.20669062435626984,
      "rewards/rejected": -0.3170241713523865,
      "step": 120
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.904486005914027e-07,
      "logits/chosen": -2.639065742492676,
      "logits/rejected": -2.5192837715148926,
      "logps/chosen": -312.64453125,
      "logps/rejected": -290.57025146484375,
      "loss": 0.6221,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.15254992246627808,
      "rewards/margins": 0.21097925305366516,
      "rewards/rejected": -0.36352917551994324,
      "step": 130
    },
    {
      "epoch": 0.88,
      "learning_rate": 2.1464952759020857e-07,
      "logits/chosen": -2.5287649631500244,
      "logits/rejected": -2.4844765663146973,
      "logps/chosen": -285.4928283691406,
      "logps/rejected": -300.5223693847656,
      "loss": 0.6143,
      "rewards/accuracies": 0.6781250238418579,
      "rewards/chosen": -0.14462201297283173,
      "rewards/margins": 0.2161392718553543,
      "rewards/rejected": -0.3607613146305084,
      "step": 140
    },
    {
      "epoch": 0.94,
      "learning_rate": 4.870879364444109e-08,
      "logits/chosen": -2.6168649196624756,
      "logits/rejected": -2.5214154720306396,
      "logps/chosen": -299.5126037597656,
      "logps/rejected": -298.0356750488281,
      "loss": 0.6145,
      "rewards/accuracies": 0.703125,
      "rewards/chosen": -0.1232418566942215,
      "rewards/margins": 0.22777679562568665,
      "rewards/rejected": -0.3510186970233917,
      "step": 150
    },
    {
      "epoch": 1.0,
      "step": 159,
      "total_flos": 0.0,
      "train_loss": 0.6436299775381508,
      "train_runtime": 7284.0023,
      "train_samples_per_second": 2.798,
      "train_steps_per_second": 0.022
    }
  ],
  "logging_steps": 10,
  "max_steps": 159,
  "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
}