{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 100,
  "global_step": 478,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0020920502092050207,
      "grad_norm": 11.329232130912374,
      "learning_rate": 1.0416666666666666e-08,
      "logits/chosen": -0.28515625,
      "logits/rejected": -0.34765625,
      "logps/chosen": -372.0,
      "logps/rejected": -440.0,
      "loss": 1.3828,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.02092050209205021,
      "grad_norm": 10.768270205160954,
      "learning_rate": 1.0416666666666667e-07,
      "logits/chosen": -0.271484375,
      "logits/rejected": -0.220703125,
      "logps/chosen": -320.0,
      "logps/rejected": -310.0,
      "loss": 1.3853,
      "rewards/accuracies": 0.2152777761220932,
      "rewards/chosen": 0.000476837158203125,
      "rewards/margins": 0.0002956390380859375,
      "rewards/rejected": 0.0001850128173828125,
      "step": 10
    },
    {
      "epoch": 0.04184100418410042,
      "grad_norm": 9.438876866649906,
      "learning_rate": 2.0833333333333333e-07,
      "logits/chosen": -0.2109375,
      "logits/rejected": -0.189453125,
      "logps/chosen": -342.0,
      "logps/rejected": -366.0,
      "loss": 1.3843,
      "rewards/accuracies": 0.23749999701976776,
      "rewards/chosen": -0.0003910064697265625,
      "rewards/margins": -0.000263214111328125,
      "rewards/rejected": -0.00012493133544921875,
      "step": 20
    },
    {
      "epoch": 0.06276150627615062,
      "grad_norm": 10.93659663876472,
      "learning_rate": 3.1249999999999997e-07,
      "logits/chosen": -0.2421875,
      "logits/rejected": -0.2060546875,
      "logps/chosen": -352.0,
      "logps/rejected": -328.0,
      "loss": 1.3813,
      "rewards/accuracies": 0.2874999940395355,
      "rewards/chosen": 0.00286865234375,
      "rewards/margins": 0.000423431396484375,
      "rewards/rejected": 0.00244140625,
      "step": 30
    },
    {
      "epoch": 0.08368200836820083,
      "grad_norm": 9.676059177655553,
      "learning_rate": 4.1666666666666667e-07,
      "logits/chosen": -0.1845703125,
      "logits/rejected": -0.1494140625,
      "logps/chosen": -314.0,
      "logps/rejected": -320.0,
      "loss": 1.3725,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": 0.0135498046875,
      "rewards/margins": 0.0135498046875,
      "rewards/rejected": -3.0547380447387695e-06,
      "step": 40
    },
    {
      "epoch": 0.10460251046025104,
      "grad_norm": 9.798546885731437,
      "learning_rate": 4.999733114418725e-07,
      "logits/chosen": -0.20703125,
      "logits/rejected": -0.1689453125,
      "logps/chosen": -362.0,
      "logps/rejected": -364.0,
      "loss": 1.3524,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": 0.037841796875,
      "rewards/margins": 0.033203125,
      "rewards/rejected": 0.0047607421875,
      "step": 50
    },
    {
      "epoch": 0.12552301255230125,
      "grad_norm": 9.60865688194641,
      "learning_rate": 4.990398100856366e-07,
      "logits/chosen": -0.2041015625,
      "logits/rejected": -0.19140625,
      "logps/chosen": -320.0,
      "logps/rejected": -366.0,
      "loss": 1.3346,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": 0.056884765625,
      "rewards/margins": 0.053466796875,
      "rewards/rejected": 0.003448486328125,
      "step": 60
    },
    {
      "epoch": 0.14644351464435146,
      "grad_norm": 9.988151072194361,
      "learning_rate": 4.967775735898179e-07,
      "logits/chosen": -0.06103515625,
      "logits/rejected": -0.068359375,
      "logps/chosen": -316.0,
      "logps/rejected": -334.0,
      "loss": 1.2953,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": 0.06787109375,
      "rewards/margins": 0.11376953125,
      "rewards/rejected": -0.0458984375,
      "step": 70
    },
    {
      "epoch": 0.16736401673640167,
      "grad_norm": 12.035906986217713,
      "learning_rate": 4.931986719649298e-07,
      "logits/chosen": -0.1591796875,
      "logits/rejected": -0.12255859375,
      "logps/chosen": -384.0,
      "logps/rejected": -340.0,
      "loss": 1.272,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": 0.050537109375,
      "rewards/margins": 0.1533203125,
      "rewards/rejected": -0.1025390625,
      "step": 80
    },
    {
      "epoch": 0.18828451882845187,
      "grad_norm": 11.972506176579511,
      "learning_rate": 4.883222001996351e-07,
      "logits/chosen": -0.12060546875,
      "logits/rejected": -0.11865234375,
      "logps/chosen": -328.0,
      "logps/rejected": -352.0,
      "loss": 1.2271,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -0.03955078125,
      "rewards/margins": 0.22265625,
      "rewards/rejected": -0.26171875,
      "step": 90
    },
    {
      "epoch": 0.20920502092050208,
      "grad_norm": 11.425153872085387,
      "learning_rate": 4.821741763807186e-07,
      "logits/chosen": -0.2119140625,
      "logits/rejected": -0.23046875,
      "logps/chosen": -340.0,
      "logps/rejected": -366.0,
      "loss": 1.2223,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.1083984375,
      "rewards/margins": 0.27734375,
      "rewards/rejected": -0.38671875,
      "step": 100
    },
    {
      "epoch": 0.20920502092050208,
      "eval_logits/chosen": -0.205078125,
      "eval_logits/rejected": -0.150390625,
      "eval_logps/chosen": -346.0,
      "eval_logps/rejected": -390.0,
      "eval_loss": 0.6163906455039978,
      "eval_rewards/accuracies": 0.734375,
      "eval_rewards/chosen": -0.1064453125,
      "eval_rewards/margins": 0.318359375,
      "eval_rewards/rejected": -0.42578125,
      "eval_runtime": 44.8743,
      "eval_samples_per_second": 44.569,
      "eval_steps_per_second": 0.713,
      "step": 100
    },
    {
      "epoch": 0.2301255230125523,
      "grad_norm": 17.93262654003255,
      "learning_rate": 4.747874028753375e-07,
      "logits/chosen": -0.240234375,
      "logits/rejected": -0.2177734375,
      "logps/chosen": -402.0,
      "logps/rejected": -378.0,
      "loss": 1.2237,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.2060546875,
      "rewards/margins": 0.2412109375,
      "rewards/rejected": -0.447265625,
      "step": 110
    },
    {
      "epoch": 0.2510460251046025,
      "grad_norm": 14.2611531769557,
      "learning_rate": 4.662012913161997e-07,
      "logits/chosen": -0.244140625,
      "logits/rejected": -0.265625,
      "logps/chosen": -392.0,
      "logps/rejected": -394.0,
      "loss": 1.193,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -0.298828125,
      "rewards/margins": 0.28125,
      "rewards/rejected": -0.58203125,
      "step": 120
    },
    {
      "epoch": 0.2719665271966527,
      "grad_norm": 12.798893628522647,
      "learning_rate": 4.5646165232345103e-07,
      "logits/chosen": -0.251953125,
      "logits/rejected": -0.2490234375,
      "logps/chosen": -358.0,
      "logps/rejected": -396.0,
      "loss": 1.18,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -0.2158203125,
      "rewards/margins": 0.359375,
      "rewards/rejected": -0.578125,
      "step": 130
    },
    {
      "epoch": 0.2928870292887029,
      "grad_norm": 19.239717506879277,
      "learning_rate": 4.456204510851956e-07,
      "logits/chosen": -0.29296875,
      "logits/rejected": -0.263671875,
      "logps/chosen": -400.0,
      "logps/rejected": -428.0,
      "loss": 1.1792,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.2734375,
      "rewards/margins": 0.37890625,
      "rewards/rejected": -0.65234375,
      "step": 140
    },
    {
      "epoch": 0.3138075313807531,
      "grad_norm": 18.538208596251096,
      "learning_rate": 4.337355301007335e-07,
      "logits/chosen": -0.287109375,
      "logits/rejected": -0.259765625,
      "logps/chosen": -380.0,
      "logps/rejected": -414.0,
      "loss": 1.1953,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.3359375,
      "rewards/margins": 0.26953125,
      "rewards/rejected": -0.60546875,
      "step": 150
    },
    {
      "epoch": 0.33472803347280333,
      "grad_norm": 13.726718415863484,
      "learning_rate": 4.2087030056579986e-07,
      "logits/chosen": -0.341796875,
      "logits/rejected": -0.345703125,
      "logps/chosen": -364.0,
      "logps/rejected": -380.0,
      "loss": 1.1963,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.2890625,
      "rewards/margins": 0.373046875,
      "rewards/rejected": -0.66015625,
      "step": 160
    },
    {
      "epoch": 0.35564853556485354,
      "grad_norm": 15.071774554309869,
      "learning_rate": 4.070934040463998e-07,
      "logits/chosen": -0.2490234375,
      "logits/rejected": -0.2265625,
      "logps/chosen": -346.0,
      "logps/rejected": -362.0,
      "loss": 1.1709,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.34765625,
      "rewards/margins": 0.31640625,
      "rewards/rejected": -0.6640625,
      "step": 170
    },
    {
      "epoch": 0.37656903765690375,
      "grad_norm": 14.777299895817363,
      "learning_rate": 3.9247834624635404e-07,
      "logits/chosen": -0.345703125,
      "logits/rejected": -0.35546875,
      "logps/chosen": -358.0,
      "logps/rejected": -368.0,
      "loss": 1.1282,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.359375,
      "rewards/margins": 0.326171875,
      "rewards/rejected": -0.68359375,
      "step": 180
    },
    {
      "epoch": 0.39748953974895396,
      "grad_norm": 17.917232829449304,
      "learning_rate": 3.7710310482256523e-07,
      "logits/chosen": -0.373046875,
      "logits/rejected": -0.3359375,
      "logps/chosen": -376.0,
      "logps/rejected": -428.0,
      "loss": 1.1659,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.30859375,
      "rewards/margins": 0.482421875,
      "rewards/rejected": -0.7890625,
      "step": 190
    },
    {
      "epoch": 0.41841004184100417,
      "grad_norm": 26.129357821594134,
      "learning_rate": 3.610497133404795e-07,
      "logits/chosen": -0.328125,
      "logits/rejected": -0.40234375,
      "logps/chosen": -370.0,
      "logps/rejected": -396.0,
      "loss": 1.1624,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": -0.37890625,
      "rewards/margins": 0.40625,
      "rewards/rejected": -0.78515625,
      "step": 200
    },
    {
      "epoch": 0.41841004184100417,
      "eval_logits/chosen": -0.400390625,
      "eval_logits/rejected": -0.36328125,
      "eval_logps/chosen": -364.0,
      "eval_logps/rejected": -430.0,
      "eval_loss": 0.5701406002044678,
      "eval_rewards/accuracies": 0.7578125,
      "eval_rewards/chosen": -0.271484375,
      "eval_rewards/margins": 0.5546875,
      "eval_rewards/rejected": -0.828125,
      "eval_runtime": 44.3219,
      "eval_samples_per_second": 45.124,
      "eval_steps_per_second": 0.722,
      "step": 200
    },
    {
      "epoch": 0.4393305439330544,
      "grad_norm": 20.166611438012268,
      "learning_rate": 3.4440382358952115e-07,
      "logits/chosen": -0.44140625,
      "logits/rejected": -0.427734375,
      "logps/chosen": -388.0,
      "logps/rejected": -406.0,
      "loss": 1.166,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -0.333984375,
      "rewards/margins": 0.462890625,
      "rewards/rejected": -0.796875,
      "step": 210
    },
    {
      "epoch": 0.4602510460251046,
      "grad_norm": 20.143454486690988,
      "learning_rate": 3.272542485937368e-07,
      "logits/chosen": -0.43359375,
      "logits/rejected": -0.4921875,
      "logps/chosen": -366.0,
      "logps/rejected": -408.0,
      "loss": 1.1348,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.37890625,
      "rewards/margins": 0.50390625,
      "rewards/rejected": -0.8828125,
      "step": 220
    },
    {
      "epoch": 0.4811715481171548,
      "grad_norm": 21.218520645945883,
      "learning_rate": 3.096924887558854e-07,
      "logits/chosen": -0.3515625,
      "logits/rejected": -0.404296875,
      "logps/chosen": -352.0,
      "logps/rejected": -408.0,
      "loss": 1.1448,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -0.443359375,
      "rewards/margins": 0.515625,
      "rewards/rejected": -0.9609375,
      "step": 230
    },
    {
      "epoch": 0.502092050209205,
      "grad_norm": 16.293281058848866,
      "learning_rate": 2.9181224366319943e-07,
      "logits/chosen": -0.3828125,
      "logits/rejected": -0.333984375,
      "logps/chosen": -366.0,
      "logps/rejected": -414.0,
      "loss": 1.1016,
      "rewards/accuracies": 0.78125,
      "rewards/chosen": -0.375,
      "rewards/margins": 0.51171875,
      "rewards/rejected": -0.88671875,
      "step": 240
    },
    {
      "epoch": 0.5230125523012552,
      "grad_norm": 14.934923417125477,
      "learning_rate": 2.7370891215954565e-07,
      "logits/chosen": -0.470703125,
      "logits/rejected": -0.40234375,
      "logps/chosen": -386.0,
      "logps/rejected": -422.0,
      "loss": 1.1298,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -0.310546875,
      "rewards/margins": 0.58203125,
      "rewards/rejected": -0.890625,
      "step": 250
    },
    {
      "epoch": 0.5439330543933054,
      "grad_norm": 17.722282481417537,
      "learning_rate": 2.55479083351317e-07,
      "logits/chosen": -0.5,
      "logits/rejected": -0.48046875,
      "logps/chosen": -392.0,
      "logps/rejected": -418.0,
      "loss": 1.1069,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": -0.376953125,
      "rewards/margins": 0.58203125,
      "rewards/rejected": -0.9609375,
      "step": 260
    },
    {
      "epoch": 0.5648535564853556,
      "grad_norm": 15.709708500953589,
      "learning_rate": 2.3722002126275822e-07,
      "logits/chosen": -0.404296875,
      "logits/rejected": -0.40625,
      "logps/chosen": -394.0,
      "logps/rejected": -414.0,
      "loss": 1.1279,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.423828125,
      "rewards/margins": 0.427734375,
      "rewards/rejected": -0.8515625,
      "step": 270
    },
    {
      "epoch": 0.5857740585774058,
      "grad_norm": 22.495247525440202,
      "learning_rate": 2.19029145890313e-07,
      "logits/chosen": -0.36328125,
      "logits/rejected": -0.37109375,
      "logps/chosen": -370.0,
      "logps/rejected": -402.0,
      "loss": 1.1459,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.369140625,
      "rewards/margins": 0.51171875,
      "rewards/rejected": -0.8828125,
      "step": 280
    },
    {
      "epoch": 0.606694560669456,
      "grad_norm": 17.685639076240154,
      "learning_rate": 2.0100351342479216e-07,
      "logits/chosen": -0.5,
      "logits/rejected": -0.423828125,
      "logps/chosen": -352.0,
      "logps/rejected": -400.0,
      "loss": 1.1235,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.412109375,
      "rewards/margins": 0.474609375,
      "rewards/rejected": -0.88671875,
      "step": 290
    },
    {
      "epoch": 0.6276150627615062,
      "grad_norm": 15.591635531747432,
      "learning_rate": 1.8323929841460178e-07,
      "logits/chosen": -0.466796875,
      "logits/rejected": -0.470703125,
      "logps/chosen": -418.0,
      "logps/rejected": -432.0,
      "loss": 1.0945,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.5,
      "rewards/margins": 0.47265625,
      "rewards/rejected": -0.97265625,
      "step": 300
    },
    {
      "epoch": 0.6276150627615062,
      "eval_logits/chosen": -0.453125,
      "eval_logits/rejected": -0.427734375,
      "eval_logps/chosen": -368.0,
      "eval_logps/rejected": -444.0,
      "eval_loss": 0.552929699420929,
      "eval_rewards/accuracies": 0.765625,
      "eval_rewards/chosen": -0.318359375,
      "eval_rewards/margins": 0.65234375,
      "eval_rewards/rejected": -0.96875,
      "eval_runtime": 44.2434,
      "eval_samples_per_second": 45.204,
      "eval_steps_per_second": 0.723,
      "step": 300
    },
    {
      "epoch": 0.6485355648535565,
      "grad_norm": 19.76547371178574,
      "learning_rate": 1.6583128063291573e-07,
      "logits/chosen": -0.341796875,
      "logits/rejected": -0.421875,
      "logps/chosen": -408.0,
      "logps/rejected": -428.0,
      "loss": 1.0912,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -0.47265625,
      "rewards/margins": 0.55078125,
      "rewards/rejected": -1.0234375,
      "step": 310
    },
    {
      "epoch": 0.6694560669456067,
      "grad_norm": 21.414058995151994,
      "learning_rate": 1.488723393865766e-07,
      "logits/chosen": -0.482421875,
      "logits/rejected": -0.4609375,
      "logps/chosen": -400.0,
      "logps/rejected": -422.0,
      "loss": 1.0604,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -0.416015625,
      "rewards/margins": 0.62890625,
      "rewards/rejected": -1.046875,
      "step": 320
    },
    {
      "epoch": 0.6903765690376569,
      "grad_norm": 19.414772003817266,
      "learning_rate": 1.3245295796480788e-07,
      "logits/chosen": -0.51171875,
      "logits/rejected": -0.498046875,
      "logps/chosen": -374.0,
      "logps/rejected": -430.0,
      "loss": 1.1006,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.46484375,
      "rewards/margins": 0.451171875,
      "rewards/rejected": -0.9140625,
      "step": 330
    },
    {
      "epoch": 0.7112970711297071,
      "grad_norm": 21.720939162926207,
      "learning_rate": 1.1666074087171627e-07,
      "logits/chosen": -0.52734375,
      "logits/rejected": -0.5078125,
      "logps/chosen": -408.0,
      "logps/rejected": -472.0,
      "loss": 1.0815,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/chosen": -0.474609375,
      "rewards/margins": 0.70703125,
      "rewards/rejected": -1.1796875,
      "step": 340
    },
    {
      "epoch": 0.7322175732217573,
      "grad_norm": 20.08927723881005,
      "learning_rate": 1.0157994641835734e-07,
      "logits/chosen": -0.45703125,
      "logits/rejected": -0.474609375,
      "logps/chosen": -376.0,
      "logps/rejected": -416.0,
      "loss": 1.0618,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.462890625,
      "rewards/margins": 0.62109375,
      "rewards/rejected": -1.0859375,
      "step": 350
    },
    {
      "epoch": 0.7531380753138075,
      "grad_norm": 19.29218318311807,
      "learning_rate": 8.729103716819111e-08,
      "logits/chosen": -0.515625,
      "logits/rejected": -0.4453125,
      "logps/chosen": -424.0,
      "logps/rejected": -448.0,
      "loss": 1.1298,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.53515625,
      "rewards/margins": 0.60546875,
      "rewards/rejected": -1.140625,
      "step": 360
    },
    {
      "epoch": 0.7740585774058577,
      "grad_norm": 18.524290784854816,
      "learning_rate": 7.387025063449081e-08,
      "logits/chosen": -0.48046875,
      "logits/rejected": -0.50390625,
      "logps/chosen": -390.0,
      "logps/rejected": -406.0,
      "loss": 1.0897,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -0.5390625,
      "rewards/margins": 0.53125,
      "rewards/rejected": -1.0703125,
      "step": 370
    },
    {
      "epoch": 0.7949790794979079,
      "grad_norm": 17.245854433432992,
      "learning_rate": 6.138919252022435e-08,
      "logits/chosen": -0.392578125,
      "logits/rejected": -0.404296875,
      "logps/chosen": -360.0,
      "logps/rejected": -456.0,
      "loss": 1.0967,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -0.5703125,
      "rewards/margins": 0.7109375,
      "rewards/rejected": -1.28125,
      "step": 380
    },
    {
      "epoch": 0.8158995815899581,
      "grad_norm": 25.69229657215138,
      "learning_rate": 4.991445467064689e-08,
      "logits/chosen": -0.51171875,
      "logits/rejected": -0.4609375,
      "logps/chosen": -424.0,
      "logps/rejected": -474.0,
      "loss": 1.0782,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.53515625,
      "rewards/margins": 0.609375,
      "rewards/rejected": -1.1484375,
      "step": 390
    },
    {
      "epoch": 0.8368200836820083,
      "grad_norm": 34.70688710928117,
      "learning_rate": 3.9507259776993954e-08,
      "logits/chosen": -0.51953125,
      "logits/rejected": -0.466796875,
      "logps/chosen": -390.0,
      "logps/rejected": -456.0,
      "loss": 1.0971,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -0.5234375,
      "rewards/margins": 0.6328125,
      "rewards/rejected": -1.15625,
      "step": 400
    },
    {
      "epoch": 0.8368200836820083,
      "eval_logits/chosen": -0.4921875,
      "eval_logits/rejected": -0.47265625,
      "eval_logps/chosen": -376.0,
      "eval_logps/rejected": -460.0,
      "eval_loss": 0.5477031469345093,
      "eval_rewards/accuracies": 0.75390625,
      "eval_rewards/chosen": -0.408203125,
      "eval_rewards/margins": 0.7265625,
      "eval_rewards/rejected": -1.1328125,
      "eval_runtime": 44.1617,
      "eval_samples_per_second": 45.288,
      "eval_steps_per_second": 0.725,
      "step": 400
    },
    {
      "epoch": 0.8577405857740585,
      "grad_norm": 19.643118113308155,
      "learning_rate": 3.022313472693447e-08,
      "logits/chosen": -0.515625,
      "logits/rejected": -0.515625,
      "logps/chosen": -398.0,
      "logps/rejected": -442.0,
      "loss": 1.101,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.47265625,
      "rewards/margins": 0.578125,
      "rewards/rejected": -1.0546875,
      "step": 410
    },
    {
      "epoch": 0.8786610878661087,
      "grad_norm": 19.28458618388238,
      "learning_rate": 2.2111614344599684e-08,
      "logits/chosen": -0.4453125,
      "logits/rejected": -0.498046875,
      "logps/chosen": -438.0,
      "logps/rejected": -480.0,
      "loss": 1.0719,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.52734375,
      "rewards/margins": 0.6015625,
      "rewards/rejected": -1.1328125,
      "step": 420
    },
    {
      "epoch": 0.899581589958159,
      "grad_norm": 16.558119542768463,
      "learning_rate": 1.521597710086439e-08,
      "logits/chosen": -0.4453125,
      "logits/rejected": -0.408203125,
      "logps/chosen": -416.0,
      "logps/rejected": -444.0,
      "loss": 1.0792,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -0.5859375,
      "rewards/margins": 0.58203125,
      "rewards/rejected": -1.1640625,
      "step": 430
    },
    {
      "epoch": 0.9205020920502092,
      "grad_norm": 19.88416183093778,
      "learning_rate": 9.57301420397924e-09,
      "logits/chosen": -0.466796875,
      "logits/rejected": -0.462890625,
      "logps/chosen": -404.0,
      "logps/rejected": -440.0,
      "loss": 1.097,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.5,
      "rewards/margins": 0.515625,
      "rewards/rejected": -1.015625,
      "step": 440
    },
    {
      "epoch": 0.9414225941422594,
      "grad_norm": 16.17448723735275,
      "learning_rate": 5.212833302556258e-09,
      "logits/chosen": -0.546875,
      "logits/rejected": -0.5625,
      "logps/chosen": -400.0,
      "logps/rejected": -486.0,
      "loss": 1.0946,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.52734375,
      "rewards/margins": 0.5625,
      "rewards/rejected": -1.0859375,
      "step": 450
    },
    {
      "epoch": 0.9623430962343096,
      "grad_norm": 25.24493230543812,
      "learning_rate": 2.158697848236607e-09,
      "logits/chosen": -0.46875,
      "logits/rejected": -0.5,
      "logps/chosen": -386.0,
      "logps/rejected": -414.0,
      "loss": 1.0964,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -0.447265625,
      "rewards/margins": 0.5546875,
      "rewards/rejected": -1.0,
      "step": 460
    },
    {
      "epoch": 0.9832635983263598,
      "grad_norm": 17.43448970223912,
      "learning_rate": 4.269029751107489e-10,
      "logits/chosen": -0.439453125,
      "logits/rejected": -0.49609375,
      "logps/chosen": -396.0,
      "logps/rejected": -456.0,
      "loss": 1.0763,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.50390625,
      "rewards/margins": 0.515625,
      "rewards/rejected": -1.0234375,
      "step": 470
    },
    {
      "epoch": 1.0,
      "step": 478,
      "total_flos": 0.0,
      "train_loss": 1.1650768583289748,
      "train_runtime": 3769.2146,
      "train_samples_per_second": 16.219,
      "train_steps_per_second": 0.127
    }
  ],
  "logging_steps": 10,
  "max_steps": 478,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}