{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9158878504672896,
  "eval_steps": 50,
  "global_step": 78,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18691588785046728,
      "grad_norm": 61.21687204105591,
      "learning_rate": 5e-07,
      "logits/chosen": -2.7639615535736084,
      "logits/rejected": -2.753985643386841,
      "logps/chosen": -265.36212158203125,
      "logps/rejected": -219.6597900390625,
      "loss": 0.6903,
      "rewards/accuracies": 0.35624998807907104,
      "rewards/chosen": 0.01777561381459236,
      "rewards/margins": 0.004832454025745392,
      "rewards/rejected": 0.012943158857524395,
      "step": 5
    },
    {
      "epoch": 0.37383177570093457,
      "grad_norm": 46.174004471862034,
      "learning_rate": 1e-06,
      "logits/chosen": -2.7134578227996826,
      "logits/rejected": -2.718207597732544,
      "logps/chosen": -287.67529296875,
      "logps/rejected": -230.2861785888672,
      "loss": 0.6281,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": 0.6617986559867859,
      "rewards/margins": 0.29742884635925293,
      "rewards/rejected": 0.36436980962753296,
      "step": 10
    },
    {
      "epoch": 0.5607476635514018,
      "grad_norm": 46.436393575993726,
      "learning_rate": 9.867190271803463e-07,
      "logits/chosen": -2.5830750465393066,
      "logits/rejected": -2.587627410888672,
      "logps/chosen": -273.77752685546875,
      "logps/rejected": -220.65695190429688,
      "loss": 0.6333,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": 1.6920897960662842,
      "rewards/margins": 0.8979904055595398,
      "rewards/rejected": 0.7940995097160339,
      "step": 15
    },
    {
      "epoch": 0.7476635514018691,
      "grad_norm": 43.837511462770195,
      "learning_rate": 9.475816456775312e-07,
      "logits/chosen": -2.471787929534912,
      "logits/rejected": -2.478492259979248,
      "logps/chosen": -246.120361328125,
      "logps/rejected": -214.670654296875,
      "loss": 0.5943,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": 1.5776749849319458,
      "rewards/margins": 1.0568134784698486,
      "rewards/rejected": 0.5208614468574524,
      "step": 20
    },
    {
      "epoch": 0.9345794392523364,
      "grad_norm": 37.09162689266799,
      "learning_rate": 8.846669854914395e-07,
      "logits/chosen": -2.4390480518341064,
      "logits/rejected": -2.411414861679077,
      "logps/chosen": -269.1220397949219,
      "logps/rejected": -211.203369140625,
      "loss": 0.5775,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": 1.8889753818511963,
      "rewards/margins": 1.2695468664169312,
      "rewards/rejected": 0.619428277015686,
      "step": 25
    },
    {
      "epoch": 1.1214953271028036,
      "grad_norm": 23.14113734554105,
      "learning_rate": 8.013173181896282e-07,
      "logits/chosen": -2.3594138622283936,
      "logits/rejected": -2.3916964530944824,
      "logps/chosen": -255.5403289794922,
      "logps/rejected": -206.4452667236328,
      "loss": 0.4246,
      "rewards/accuracies": 0.8500000238418579,
      "rewards/chosen": 2.106290340423584,
      "rewards/margins": 2.024784564971924,
      "rewards/rejected": 0.08150559663772583,
      "step": 30
    },
    {
      "epoch": 1.308411214953271,
      "grad_norm": 20.944832198946614,
      "learning_rate": 7.019605024359474e-07,
      "logits/chosen": -2.3514552116394043,
      "logits/rejected": -2.332791805267334,
      "logps/chosen": -238.28518676757812,
      "logps/rejected": -240.40560913085938,
      "loss": 0.2842,
      "rewards/accuracies": 0.875,
      "rewards/chosen": 2.428344249725342,
      "rewards/margins": 3.228165864944458,
      "rewards/rejected": -0.7998219728469849,
      "step": 35
    },
    {
      "epoch": 1.4953271028037383,
      "grad_norm": 30.791834420795908,
      "learning_rate": 5.918747589082852e-07,
      "logits/chosen": -2.33073353767395,
      "logits/rejected": -2.3257832527160645,
      "logps/chosen": -250.0792694091797,
      "logps/rejected": -221.3015594482422,
      "loss": 0.2477,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": 2.562706232070923,
      "rewards/margins": 3.5494167804718018,
      "rewards/rejected": -0.9867107272148132,
      "step": 40
    },
    {
      "epoch": 1.6822429906542056,
      "grad_norm": 27.339770904226572,
      "learning_rate": 4.769082706771303e-07,
      "logits/chosen": -2.354419708251953,
      "logits/rejected": -2.345761775970459,
      "logps/chosen": -251.4537811279297,
      "logps/rejected": -224.0857391357422,
      "loss": 0.282,
      "rewards/accuracies": 0.8500000238418579,
      "rewards/chosen": 2.958096742630005,
      "rewards/margins": 3.320951461791992,
      "rewards/rejected": -0.3628546893596649,
      "step": 45
    },
    {
      "epoch": 1.8691588785046729,
      "grad_norm": 26.19581986930712,
      "learning_rate": 3.6316850496395855e-07,
      "logits/chosen": -2.395775318145752,
      "logits/rejected": -2.3773467540740967,
      "logps/chosen": -227.96786499023438,
      "logps/rejected": -228.518310546875,
      "loss": 0.3348,
      "rewards/accuracies": 0.8187500238418579,
      "rewards/chosen": 2.7191131114959717,
      "rewards/margins": 2.9623184204101562,
      "rewards/rejected": -0.24320510029792786,
      "step": 50
    },
    {
      "epoch": 1.8691588785046729,
      "eval_logits/chosen": -2.406548023223877,
      "eval_logits/rejected": -2.4020678997039795,
      "eval_logps/chosen": -249.0997314453125,
      "eval_logps/rejected": -190.4173583984375,
      "eval_loss": 0.5792086124420166,
      "eval_rewards/accuracies": 0.8125,
      "eval_rewards/chosen": 2.028245210647583,
      "eval_rewards/margins": 1.727249026298523,
      "eval_rewards/rejected": 0.300996333360672,
      "eval_runtime": 50.8966,
      "eval_samples_per_second": 14.932,
      "eval_steps_per_second": 0.236,
      "step": 50
    },
    {
      "epoch": 2.05607476635514,
      "grad_norm": 19.061520787069963,
      "learning_rate": 2.566977607165719e-07,
      "logits/chosen": -2.425044298171997,
      "logits/rejected": -2.430022716522217,
      "logps/chosen": -258.14617919921875,
      "logps/rejected": -232.40603637695312,
      "loss": 0.2745,
      "rewards/accuracies": 0.90625,
      "rewards/chosen": 2.87212872505188,
      "rewards/margins": 3.6639626026153564,
      "rewards/rejected": -0.7918335795402527,
      "step": 55
    },
    {
      "epoch": 2.2429906542056073,
      "grad_norm": 15.460701798223598,
      "learning_rate": 1.631521781767214e-07,
      "logits/chosen": -2.4355878829956055,
      "logits/rejected": -2.4206368923187256,
      "logps/chosen": -223.0980987548828,
      "logps/rejected": -222.94241333007812,
      "loss": 0.1636,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": 2.6068944931030273,
      "rewards/margins": 3.432859420776367,
      "rewards/rejected": -0.8259647488594055,
      "step": 60
    },
    {
      "epoch": 2.4299065420560746,
      "grad_norm": 16.614589190551243,
      "learning_rate": 8.75012627008489e-08,
      "logits/chosen": -2.4444804191589355,
      "logits/rejected": -2.4162566661834717,
      "logps/chosen": -244.12841796875,
      "logps/rejected": -229.2833251953125,
      "loss": 0.1617,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": 2.8097472190856934,
      "rewards/margins": 3.784661054611206,
      "rewards/rejected": -0.9749139547348022,
      "step": 65
    },
    {
      "epoch": 2.616822429906542,
      "grad_norm": 19.056173520446013,
      "learning_rate": 3.376388529782215e-08,
      "logits/chosen": -2.4610393047332764,
      "logits/rejected": -2.440760612487793,
      "logps/chosen": -251.9339599609375,
      "logps/rejected": -247.04275512695312,
      "loss": 0.1661,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": 3.2026991844177246,
      "rewards/margins": 4.342484474182129,
      "rewards/rejected": -1.1397849321365356,
      "step": 70
    },
    {
      "epoch": 2.803738317757009,
      "grad_norm": 22.14435014976595,
      "learning_rate": 4.794784562397458e-09,
      "logits/chosen": -2.439197063446045,
      "logits/rejected": -2.4480080604553223,
      "logps/chosen": -245.83792114257812,
      "logps/rejected": -227.4940185546875,
      "loss": 0.1707,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": 3.1310062408447266,
      "rewards/margins": 3.876368761062622,
      "rewards/rejected": -0.7453619241714478,
      "step": 75
    }
  ],
  "logging_steps": 5,
  "max_steps": 78,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 919378820333568.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}