{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984,
  "eval_steps": 200,
  "global_step": 312,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0032,
      "grad_norm": 309.07493134536526,
      "learning_rate": 1.5625e-08,
      "logits/generated": -3.1874351501464844,
      "logits/real": -2.811344623565674,
      "logps/generated": -277.39678955078125,
      "logps/real": -164.29153442382812,
      "loss": 0.8248,
      "rewards/accuracies": 0.0,
      "rewards/generated": 0.0,
      "rewards/margins": 0.0,
      "rewards/real": 0.0,
      "step": 1
    },
    {
      "epoch": 0.032,
      "grad_norm": 13.846248517181778,
      "learning_rate": 1.5624999999999999e-07,
      "logits/generated": -2.9546539783477783,
      "logits/real": -2.5018956661224365,
      "logps/generated": -249.66824340820312,
      "logps/real": -122.42621612548828,
      "loss": 0.5265,
      "rewards/accuracies": 0.8888888955116272,
      "rewards/generated": -0.8262240886688232,
      "rewards/margins": 1.2902787923812866,
      "rewards/real": 0.46405479311943054,
      "step": 10
    },
    {
      "epoch": 0.064,
      "grad_norm": 1.537832573373159,
      "learning_rate": 3.1249999999999997e-07,
      "logits/generated": -3.0161662101745605,
      "logits/real": -2.225759267807007,
      "logps/generated": -312.00897216796875,
      "logps/real": -104.78621673583984,
      "loss": 0.1131,
      "rewards/accuracies": 1.0,
      "rewards/generated": -5.391150951385498,
      "rewards/margins": 8.365974426269531,
      "rewards/real": 2.9748237133026123,
      "step": 20
    },
    {
      "epoch": 0.096,
      "grad_norm": 1.5948197394392936,
      "learning_rate": 4.6874999999999996e-07,
      "logits/generated": -2.8325116634368896,
      "logits/real": -2.309696674346924,
      "logps/generated": -343.28424072265625,
      "logps/real": -100.86705780029297,
      "loss": 0.1004,
      "rewards/accuracies": 1.0,
      "rewards/generated": -8.481008529663086,
      "rewards/margins": 12.336121559143066,
      "rewards/real": 3.855113983154297,
      "step": 30
    },
    {
      "epoch": 0.128,
      "grad_norm": 1.360990457967047,
      "learning_rate": 4.857142857142857e-07,
      "logits/generated": -2.8159852027893066,
      "logits/real": -2.518409013748169,
      "logps/generated": -355.71441650390625,
      "logps/real": -102.14837646484375,
      "loss": 0.1018,
      "rewards/accuracies": 1.0,
      "rewards/generated": -10.088699340820312,
      "rewards/margins": 14.381967544555664,
      "rewards/real": 4.293266773223877,
      "step": 40
    },
    {
      "epoch": 0.16,
      "grad_norm": 1.1637353343229784,
      "learning_rate": 4.6785714285714283e-07,
      "logits/generated": -2.810716152191162,
      "logits/real": -2.397970199584961,
      "logps/generated": -361.4827575683594,
      "logps/real": -93.07586669921875,
      "loss": 0.0866,
      "rewards/accuracies": 1.0,
      "rewards/generated": -10.693376541137695,
      "rewards/margins": 14.997743606567383,
      "rewards/real": 4.304368019104004,
      "step": 50
    },
    {
      "epoch": 0.192,
      "grad_norm": 1.5949083844151817,
      "learning_rate": 4.5e-07,
      "logits/generated": -2.5569677352905273,
      "logits/real": -2.209700584411621,
      "logps/generated": -356.9215393066406,
      "logps/real": -90.63230895996094,
      "loss": 0.0871,
      "rewards/accuracies": 1.0,
      "rewards/generated": -11.640242576599121,
      "rewards/margins": 16.65045928955078,
      "rewards/real": 5.0102152824401855,
      "step": 60
    },
    {
      "epoch": 0.224,
      "grad_norm": 1.1997502879904265,
      "learning_rate": 4.3214285714285713e-07,
      "logits/generated": -2.778857946395874,
      "logits/real": -2.4418845176696777,
      "logps/generated": -363.03106689453125,
      "logps/real": -83.89906311035156,
      "loss": 0.0889,
      "rewards/accuracies": 1.0,
      "rewards/generated": -11.029804229736328,
      "rewards/margins": 16.131534576416016,
      "rewards/real": 5.1017303466796875,
      "step": 70
    },
    {
      "epoch": 0.256,
      "grad_norm": 1.2541389286949767,
      "learning_rate": 4.142857142857143e-07,
      "logits/generated": -2.748169422149658,
      "logits/real": -2.537468910217285,
      "logps/generated": -352.0079040527344,
      "logps/real": -90.74658966064453,
      "loss": 0.0847,
      "rewards/accuracies": 1.0,
      "rewards/generated": -10.95399284362793,
      "rewards/margins": 16.11164093017578,
      "rewards/real": 5.157646179199219,
      "step": 80
    },
    {
      "epoch": 0.288,
      "grad_norm": 1.3716409313987092,
      "learning_rate": 3.9642857142857137e-07,
      "logits/generated": -2.7057158946990967,
      "logits/real": -2.4590487480163574,
      "logps/generated": -369.472412109375,
      "logps/real": -93.29542541503906,
      "loss": 0.0867,
      "rewards/accuracies": 1.0,
      "rewards/generated": -11.155221939086914,
      "rewards/margins": 17.421768188476562,
      "rewards/real": 6.266547203063965,
      "step": 90
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.2776829690193363,
      "learning_rate": 3.785714285714285e-07,
      "logits/generated": -2.7385950088500977,
      "logits/real": -2.558948040008545,
      "logps/generated": -360.800048828125,
      "logps/real": -94.03065490722656,
      "loss": 0.0812,
      "rewards/accuracies": 1.0,
      "rewards/generated": -11.058977127075195,
      "rewards/margins": 17.12522315979004,
      "rewards/real": 6.066246032714844,
      "step": 100
    },
    {
      "epoch": 0.352,
      "grad_norm": 1.1502347571032556,
      "learning_rate": 3.607142857142857e-07,
      "logits/generated": -2.7538702487945557,
      "logits/real": -2.577524185180664,
      "logps/generated": -369.98388671875,
      "logps/real": -81.54554748535156,
      "loss": 0.0768,
      "rewards/accuracies": 1.0,
      "rewards/generated": -12.037736892700195,
      "rewards/margins": 18.702112197875977,
      "rewards/real": 6.664376735687256,
      "step": 110
    },
    {
      "epoch": 0.384,
      "grad_norm": 1.3343308865576284,
      "learning_rate": 3.4285714285714286e-07,
      "logits/generated": -2.6860907077789307,
      "logits/real": -2.498666763305664,
      "logps/generated": -367.09027099609375,
      "logps/real": -77.16963195800781,
      "loss": 0.0813,
      "rewards/accuracies": 1.0,
      "rewards/generated": -12.206511497497559,
      "rewards/margins": 18.975597381591797,
      "rewards/real": 6.7690839767456055,
      "step": 120
    },
    {
      "epoch": 0.416,
      "grad_norm": 1.3770646126862427,
      "learning_rate": 3.25e-07,
      "logits/generated": -2.624868869781494,
      "logits/real": -2.4215009212493896,
      "logps/generated": -371.2015380859375,
      "logps/real": -70.61559295654297,
      "loss": 0.0775,
      "rewards/accuracies": 1.0,
      "rewards/generated": -12.158758163452148,
      "rewards/margins": 18.808048248291016,
      "rewards/real": 6.649291038513184,
      "step": 130
    },
    {
      "epoch": 0.448,
      "grad_norm": 1.4072742854376892,
      "learning_rate": 3.0714285714285716e-07,
      "logits/generated": -2.632169246673584,
      "logits/real": -2.5124592781066895,
      "logps/generated": -378.91949462890625,
      "logps/real": -85.79364013671875,
      "loss": 0.0783,
      "rewards/accuracies": 1.0,
      "rewards/generated": -12.807360649108887,
      "rewards/margins": 20.574337005615234,
      "rewards/real": 7.7669782638549805,
      "step": 140
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.6731854026570883,
      "learning_rate": 2.892857142857143e-07,
      "logits/generated": -2.7312309741973877,
      "logits/real": -2.513092517852783,
      "logps/generated": -376.5494689941406,
      "logps/real": -86.8825912475586,
      "loss": 0.0719,
      "rewards/accuracies": 1.0,
      "rewards/generated": -12.506660461425781,
      "rewards/margins": 19.993375778198242,
      "rewards/real": 7.486716270446777,
      "step": 150
    },
    {
      "epoch": 0.512,
      "grad_norm": 1.7208597186189485,
      "learning_rate": 2.714285714285714e-07,
      "logits/generated": -2.6538147926330566,
      "logits/real": -2.4443583488464355,
      "logps/generated": -379.8941955566406,
      "logps/real": -63.8643798828125,
      "loss": 0.0664,
      "rewards/accuracies": 1.0,
      "rewards/generated": -12.761404991149902,
      "rewards/margins": 21.195444107055664,
      "rewards/real": 8.434039115905762,
      "step": 160
    },
    {
      "epoch": 0.544,
      "grad_norm": 3.296728367897804,
      "learning_rate": 2.5357142857142855e-07,
      "logits/generated": -2.679011821746826,
      "logits/real": -2.4271175861358643,
      "logps/generated": -368.987060546875,
      "logps/real": -65.5553207397461,
      "loss": 0.0683,
      "rewards/accuracies": 1.0,
      "rewards/generated": -12.584733963012695,
      "rewards/margins": 19.998342514038086,
      "rewards/real": 7.413609504699707,
      "step": 170
    },
    {
      "epoch": 0.576,
      "grad_norm": 1.509303126217559,
      "learning_rate": 2.357142857142857e-07,
      "logits/generated": -2.706902503967285,
      "logits/real": -2.339202404022217,
      "logps/generated": -393.6845703125,
      "logps/real": -57.10161590576172,
      "loss": 0.0613,
      "rewards/accuracies": 1.0,
      "rewards/generated": -13.54321575164795,
      "rewards/margins": 20.973468780517578,
      "rewards/real": 7.430250644683838,
      "step": 180
    },
    {
      "epoch": 0.608,
      "grad_norm": 1.6226015420004847,
      "learning_rate": 2.1785714285714284e-07,
      "logits/generated": -2.6822266578674316,
      "logits/real": -2.390279769897461,
      "logps/generated": -391.8895263671875,
      "logps/real": -64.2778549194336,
      "loss": 0.0593,
      "rewards/accuracies": 1.0,
      "rewards/generated": -13.559152603149414,
      "rewards/margins": 21.35408592224121,
      "rewards/real": 7.794934272766113,
      "step": 190
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.4621480589976257,
      "learning_rate": 2e-07,
      "logits/generated": -2.64497447013855,
      "logits/real": -2.3843231201171875,
      "logps/generated": -378.09307861328125,
      "logps/real": -53.13884353637695,
      "loss": 0.0622,
      "rewards/accuracies": 1.0,
      "rewards/generated": -12.936541557312012,
      "rewards/margins": 21.156665802001953,
      "rewards/real": 8.220122337341309,
      "step": 200
    },
    {
      "epoch": 0.64,
      "eval_logits/generated": -2.709409236907959,
      "eval_logits/real": -2.410205602645874,
      "eval_logps/generated": -384.0584411621094,
      "eval_logps/real": -105.40039825439453,
      "eval_loss": 0.10775644332170486,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/generated": -13.081828117370605,
      "eval_rewards/margins": 16.210613250732422,
      "eval_rewards/real": 3.1287851333618164,
      "eval_runtime": 31.4401,
      "eval_samples_per_second": 6.361,
      "eval_steps_per_second": 0.413,
      "step": 200
    },
    {
      "epoch": 0.672,
      "grad_norm": 1.6581201639125833,
      "learning_rate": 1.8214285714285714e-07,
      "logits/generated": -2.6259570121765137,
      "logits/real": -2.3245978355407715,
      "logps/generated": -380.79241943359375,
      "logps/real": -50.675270080566406,
      "loss": 0.0563,
      "rewards/accuracies": 1.0,
      "rewards/generated": -13.35838508605957,
      "rewards/margins": 21.13888168334961,
      "rewards/real": 7.780497074127197,
      "step": 210
    },
    {
      "epoch": 0.704,
      "grad_norm": 2.0377232989636846,
      "learning_rate": 1.6428571428571429e-07,
      "logits/generated": -2.6223349571228027,
      "logits/real": -2.4116692543029785,
      "logps/generated": -388.216064453125,
      "logps/real": -75.10206604003906,
      "loss": 0.0601,
      "rewards/accuracies": 1.0,
      "rewards/generated": -13.288862228393555,
      "rewards/margins": 22.80645751953125,
      "rewards/real": 9.51759147644043,
      "step": 220
    },
    {
      "epoch": 0.736,
      "grad_norm": 1.4117008274451168,
      "learning_rate": 1.4642857142857143e-07,
      "logits/generated": -2.6506168842315674,
      "logits/real": -2.356983184814453,
      "logps/generated": -402.24298095703125,
      "logps/real": -52.900665283203125,
      "loss": 0.051,
      "rewards/accuracies": 1.0,
      "rewards/generated": -14.113009452819824,
      "rewards/margins": 22.862171173095703,
      "rewards/real": 8.749161720275879,
      "step": 230
    },
    {
      "epoch": 0.768,
      "grad_norm": 1.5408677183120254,
      "learning_rate": 1.2857142857142855e-07,
      "logits/generated": -2.663072109222412,
      "logits/real": -2.295074939727783,
      "logps/generated": -392.62335205078125,
      "logps/real": -48.768280029296875,
      "loss": 0.0565,
      "rewards/accuracies": 1.0,
      "rewards/generated": -13.465108871459961,
      "rewards/margins": 21.69220542907715,
      "rewards/real": 8.227096557617188,
      "step": 240
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.6617128645150359,
      "learning_rate": 1.107142857142857e-07,
      "logits/generated": -2.659288167953491,
      "logits/real": -2.284973621368408,
      "logps/generated": -391.0361022949219,
      "logps/real": -49.54872131347656,
      "loss": 0.0525,
      "rewards/accuracies": 1.0,
      "rewards/generated": -13.513246536254883,
      "rewards/margins": 21.818363189697266,
      "rewards/real": 8.30511474609375,
      "step": 250
    },
    {
      "epoch": 0.832,
      "grad_norm": 1.9545108954634698,
      "learning_rate": 9.285714285714286e-08,
      "logits/generated": -2.5850462913513184,
      "logits/real": -2.2890560626983643,
      "logps/generated": -377.42095947265625,
      "logps/real": -50.45993423461914,
      "loss": 0.0541,
      "rewards/accuracies": 1.0,
      "rewards/generated": -13.420709609985352,
      "rewards/margins": 21.94915008544922,
      "rewards/real": 8.528444290161133,
      "step": 260
    },
    {
      "epoch": 0.864,
      "grad_norm": 1.663160142834365,
      "learning_rate": 7.5e-08,
      "logits/generated": -2.639012098312378,
      "logits/real": -2.241528272628784,
      "logps/generated": -385.27972412109375,
      "logps/real": -44.987178802490234,
      "loss": 0.0519,
      "rewards/accuracies": 1.0,
      "rewards/generated": -13.790578842163086,
      "rewards/margins": 21.756423950195312,
      "rewards/real": 7.965843200683594,
      "step": 270
    },
    {
      "epoch": 0.896,
      "grad_norm": 1.7475888656947995,
      "learning_rate": 5.714285714285714e-08,
      "logits/generated": -2.634594440460205,
      "logits/real": -2.309357166290283,
      "logps/generated": -383.92266845703125,
      "logps/real": -47.04390335083008,
      "loss": 0.0523,
      "rewards/accuracies": 1.0,
      "rewards/generated": -13.899473190307617,
      "rewards/margins": 22.34453582763672,
      "rewards/real": 8.445062637329102,
      "step": 280
    },
    {
      "epoch": 0.928,
      "grad_norm": 1.8843842875992995,
      "learning_rate": 3.9285714285714285e-08,
      "logits/generated": -2.6036441326141357,
      "logits/real": -2.2659499645233154,
      "logps/generated": -395.3347473144531,
      "logps/real": -55.319923400878906,
      "loss": 0.0537,
      "rewards/accuracies": 1.0,
      "rewards/generated": -14.550788879394531,
      "rewards/margins": 24.063692092895508,
      "rewards/real": 9.512903213500977,
      "step": 290
    },
    {
      "epoch": 0.96,
      "grad_norm": 1.7336216562094997,
      "learning_rate": 2.142857142857143e-08,
      "logits/generated": -2.5489935874938965,
      "logits/real": -2.3403241634368896,
      "logps/generated": -381.0558166503906,
      "logps/real": -60.32090377807617,
      "loss": 0.0515,
      "rewards/accuracies": 1.0,
      "rewards/generated": -13.906036376953125,
      "rewards/margins": 23.397653579711914,
      "rewards/real": 9.491620063781738,
      "step": 300
    },
    {
      "epoch": 0.992,
      "grad_norm": 1.7332047882804558,
      "learning_rate": 3.571428571428571e-09,
      "logits/generated": -2.6001510620117188,
      "logits/real": -2.3211655616760254,
      "logps/generated": -384.77154541015625,
      "logps/real": -53.601829528808594,
      "loss": 0.0525,
      "rewards/accuracies": 1.0,
      "rewards/generated": -13.838152885437012,
      "rewards/margins": 23.025869369506836,
      "rewards/real": 9.187714576721191,
      "step": 310
    },
    {
      "epoch": 0.9984,
      "step": 312,
      "total_flos": 0.0,
      "train_loss": 0.086211400011029,
      "train_runtime": 3213.2155,
      "train_samples_per_second": 3.111,
      "train_steps_per_second": 0.097
    }
  ],
  "logging_steps": 10,
  "max_steps": 312,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}