{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9991836734693877,
  "eval_steps": 100,
  "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-07,
      "logits/chosen": -2.8246347904205322,
      "logits/rejected": -2.7856249809265137,
      "logps/chosen": -238.8000030517578,
      "logps/rejected": -252.79095458984375,
      "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": 3.125e-06,
      "logits/chosen": -2.786532402038574,
      "logits/rejected": -2.722121238708496,
      "logps/chosen": -234.37245178222656,
      "logps/rejected": -252.76654052734375,
      "loss": 0.6931,
      "rewards/accuracies": 0.3854166567325592,
      "rewards/chosen": 0.00019371736561879516,
      "rewards/margins": -0.0003117284504696727,
      "rewards/rejected": 0.0005054458160884678,
      "step": 10
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.989490450759331e-06,
      "logits/chosen": -2.761892318725586,
      "logits/rejected": -2.7504990100860596,
      "logps/chosen": -254.9222869873047,
      "logps/rejected": -267.8934326171875,
      "loss": 0.6925,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": 0.008798873052001,
      "rewards/margins": 0.0012144726933911443,
      "rewards/rejected": 0.007584400475025177,
      "step": 20
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.872270441827174e-06,
      "logits/chosen": -2.7136857509613037,
      "logits/rejected": -2.7073874473571777,
      "logps/chosen": -241.10507202148438,
      "logps/rejected": -277.07415771484375,
      "loss": 0.6916,
      "rewards/accuracies": 0.5843750238418579,
      "rewards/chosen": 0.036142051219940186,
      "rewards/margins": 0.003209862159565091,
      "rewards/rejected": 0.032932184636592865,
      "step": 30
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.630851211353007e-06,
      "logits/chosen": -2.719532012939453,
      "logits/rejected": -2.6920456886291504,
      "logps/chosen": -238.81729125976562,
      "logps/rejected": -257.0670471191406,
      "loss": 0.6898,
      "rewards/accuracies": 0.640625,
      "rewards/chosen": 0.05626382678747177,
      "rewards/margins": 0.008078296668827534,
      "rewards/rejected": 0.048185527324676514,
      "step": 40
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.277872161641682e-06,
      "logits/chosen": -2.707075357437134,
      "logits/rejected": -2.682654619216919,
      "logps/chosen": -214.5452117919922,
      "logps/rejected": -231.02877807617188,
      "loss": 0.6873,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": 0.045605290681123734,
      "rewards/margins": 0.007303851656615734,
      "rewards/rejected": 0.038301438093185425,
      "step": 50
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.831813362428005e-06,
      "logits/chosen": -2.7438042163848877,
      "logits/rejected": -2.724104166030884,
      "logps/chosen": -239.53213500976562,
      "logps/rejected": -274.93902587890625,
      "loss": 0.6861,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": 0.010097012855112553,
      "rewards/margins": 0.017007894814014435,
      "rewards/rejected": -0.006910882890224457,
      "step": 60
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.3160280345958614e-06,
      "logits/chosen": -2.7218570709228516,
      "logits/rejected": -2.7025914192199707,
      "logps/chosen": -247.56350708007812,
      "logps/rejected": -262.78961181640625,
      "loss": 0.6839,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": 0.01939699612557888,
      "rewards/margins": 0.020038722082972527,
      "rewards/rejected": -0.0006417257827706635,
      "step": 70
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.757519902117886e-06,
      "logits/chosen": -2.650240898132324,
      "logits/rejected": -2.6454017162323,
      "logps/chosen": -229.64468383789062,
      "logps/rejected": -238.07791137695312,
      "loss": 0.6811,
      "rewards/accuracies": 0.6343749761581421,
      "rewards/chosen": -0.007218700833618641,
      "rewards/margins": 0.023447707295417786,
      "rewards/rejected": -0.030666405335068703,
      "step": 80
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.185529423440807e-06,
      "logits/chosen": -2.719679117202759,
      "logits/rejected": -2.6773438453674316,
      "logps/chosen": -231.2781524658203,
      "logps/rejected": -246.88418579101562,
      "loss": 0.6793,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.028139105066657066,
      "rewards/margins": 0.027326997369527817,
      "rewards/rejected": -0.05546610429883003,
      "step": 90
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.6300029195778454e-06,
      "logits/chosen": -2.651785373687744,
      "logits/rejected": -2.6500556468963623,
      "logps/chosen": -232.26229858398438,
      "logps/rejected": -263.81927490234375,
      "loss": 0.6793,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.013562011532485485,
      "rewards/margins": 0.03466662019491196,
      "rewards/rejected": -0.048228632658720016,
      "step": 100
    },
    {
      "epoch": 0.65,
      "eval_logits/chosen": -2.7459018230438232,
      "eval_logits/rejected": -2.6667211055755615,
      "eval_logps/chosen": -288.20477294921875,
      "eval_logps/rejected": -266.5933532714844,
      "eval_loss": 0.6749266386032104,
      "eval_rewards/accuracies": 0.6079999804496765,
      "eval_rewards/chosen": -0.04165023937821388,
      "eval_rewards/margins": 0.04912487417459488,
      "eval_rewards/rejected": -0.09077510982751846,
      "eval_runtime": 384.036,
      "eval_samples_per_second": 5.208,
      "eval_steps_per_second": 0.651,
      "step": 100
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.1200247470632394e-06,
      "logits/chosen": -2.6394271850585938,
      "logits/rejected": -2.623563766479492,
      "logps/chosen": -241.2955780029297,
      "logps/rejected": -255.7236785888672,
      "loss": 0.677,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -0.05771399661898613,
      "rewards/margins": 0.021994810551404953,
      "rewards/rejected": -0.07970880717039108,
      "step": 110
    },
    {
      "epoch": 0.78,
      "learning_rate": 6.822945986946386e-07,
      "logits/chosen": -2.7274205684661865,
      "logits/rejected": -2.681755542755127,
      "logps/chosen": -255.23196411132812,
      "logps/rejected": -279.19464111328125,
      "loss": 0.6754,
      "rewards/accuracies": 0.6031249761581421,
      "rewards/chosen": -0.05603557080030441,
      "rewards/margins": 0.03728459030389786,
      "rewards/rejected": -0.09332015365362167,
      "step": 120
    },
    {
      "epoch": 0.85,
      "learning_rate": 3.397296523427807e-07,
      "logits/chosen": -2.654305934906006,
      "logits/rejected": -2.646434783935547,
      "logps/chosen": -262.47210693359375,
      "logps/rejected": -293.465087890625,
      "loss": 0.67,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.05943986773490906,
      "rewards/margins": 0.05442862585186958,
      "rewards/rejected": -0.11386849731206894,
      "step": 130
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.102647517397798e-07,
      "logits/chosen": -2.711536407470703,
      "logits/rejected": -2.693706512451172,
      "logps/chosen": -233.05447387695312,
      "logps/rejected": -257.82183837890625,
      "loss": 0.6737,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.06369104981422424,
      "rewards/margins": 0.037472162395715714,
      "rewards/rejected": -0.10116322338581085,
      "step": 140
    },
    {
      "epoch": 0.98,
      "learning_rate": 5.9134352763748345e-09,
      "logits/chosen": -2.659524440765381,
      "logits/rejected": -2.654564380645752,
      "logps/chosen": -237.49960327148438,
      "logps/rejected": -260.6737365722656,
      "loss": 0.6724,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.05076870322227478,
      "rewards/margins": 0.045312874019145966,
      "rewards/rejected": -0.09608156979084015,
      "step": 150
    },
    {
      "epoch": 1.0,
      "step": 153,
      "total_flos": 0.0,
      "train_loss": 0.6821690501730426,
      "train_runtime": 6321.1061,
      "train_samples_per_second": 3.101,
      "train_steps_per_second": 0.024
    }
  ],
  "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
}