{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 1000,
  "global_step": 125,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "grad_norm": 0.2611100745548823,
      "learning_rate": 3.846153846153847e-07,
      "logits/chosen": -1.7501684427261353,
      "logits/rejected": -1.5946102142333984,
      "logps/chosen": -84.6928939819336,
      "logps/rejected": -66.21642303466797,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.22439873610113092,
      "learning_rate": 3.846153846153847e-06,
      "logits/chosen": -1.7436565160751343,
      "logits/rejected": -1.7198147773742676,
      "logps/chosen": -75.07618713378906,
      "logps/rejected": -78.49008178710938,
      "loss": 0.6926,
      "rewards/accuracies": 0.6944444179534912,
      "rewards/chosen": 0.0014147770125418901,
      "rewards/margins": 0.001200773986056447,
      "rewards/rejected": 0.00021400292462203652,
      "step": 10
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.23238100037617523,
      "learning_rate": 4.9519632010080765e-06,
      "logits/chosen": -1.748544454574585,
      "logits/rejected": -1.6868740320205688,
      "logps/chosen": -76.10453796386719,
      "logps/rejected": -74.7530288696289,
      "loss": 0.6845,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.013784198090434074,
      "rewards/margins": 0.017130162566900253,
      "rewards/rejected": -0.00334596517495811,
      "step": 20
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.2815020587025619,
      "learning_rate": 4.721114089947181e-06,
      "logits/chosen": -1.689690351486206,
      "logits/rejected": -1.7480777502059937,
      "logps/chosen": -69.04388427734375,
      "logps/rejected": -83.00643920898438,
      "loss": 0.6604,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.04013435170054436,
      "rewards/margins": 0.0667264312505722,
      "rewards/rejected": -0.02659207209944725,
      "step": 30
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.5012169942018447,
      "learning_rate": 4.316650805085068e-06,
      "logits/chosen": -1.6957839727401733,
      "logits/rejected": -1.7012145519256592,
      "logps/chosen": -66.49241638183594,
      "logps/rejected": -90.0240249633789,
      "loss": 0.6028,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.0827840119600296,
      "rewards/margins": 0.19734814763069153,
      "rewards/rejected": -0.11456414312124252,
      "step": 40
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.5738595020225901,
      "learning_rate": 3.770188363116324e-06,
      "logits/chosen": -1.6569753885269165,
      "logits/rejected": -1.5992107391357422,
      "logps/chosen": -59.36053466796875,
      "logps/rejected": -114.33049011230469,
      "loss": 0.4751,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.16371119022369385,
      "rewards/margins": 0.5395737886428833,
      "rewards/rejected": -0.37586259841918945,
      "step": 50
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.3468932614622257,
      "learning_rate": 3.1244411954180677e-06,
      "logits/chosen": -1.6431405544281006,
      "logits/rejected": -1.4894822835922241,
      "logps/chosen": -53.25394821166992,
      "logps/rejected": -131.90545654296875,
      "loss": 0.3531,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.2888104021549225,
      "rewards/margins": 0.8971719741821289,
      "rewards/rejected": -0.6083616018295288,
      "step": 60
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.2456344580019736,
      "learning_rate": 2.429884359310328e-06,
      "logits/chosen": -1.4915359020233154,
      "logits/rejected": -1.4094572067260742,
      "logps/chosen": -45.98115539550781,
      "logps/rejected": -179.42784118652344,
      "loss": 0.2706,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.3235810399055481,
      "rewards/margins": 1.3730812072753906,
      "rewards/rejected": -1.0495002269744873,
      "step": 70
    },
    {
      "epoch": 3.2,
      "grad_norm": 0.21353252313421264,
      "learning_rate": 1.7408081372259633e-06,
      "logits/chosen": -1.4443209171295166,
      "logits/rejected": -1.3889185190200806,
      "logps/chosen": -41.034461975097656,
      "logps/rejected": -219.5689697265625,
      "loss": 0.2546,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.3277435898780823,
      "rewards/margins": 1.7629969120025635,
      "rewards/rejected": -1.435253381729126,
      "step": 80
    },
    {
      "epoch": 3.6,
      "grad_norm": 0.147745405217318,
      "learning_rate": 1.1110744174509952e-06,
      "logits/chosen": -1.5578348636627197,
      "logits/rejected": -1.5076854228973389,
      "logps/chosen": -40.28083801269531,
      "logps/rejected": -228.35556030273438,
      "loss": 0.2351,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.3738650679588318,
      "rewards/margins": 1.9237592220306396,
      "rewards/rejected": -1.5498943328857422,
      "step": 90
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.14178606869515006,
      "learning_rate": 5.899065604459814e-07,
      "logits/chosen": -1.4096633195877075,
      "logits/rejected": -1.4153110980987549,
      "logps/chosen": -38.80024719238281,
      "logps/rejected": -268.9517822265625,
      "loss": 0.2152,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.3712707459926605,
      "rewards/margins": 2.266303300857544,
      "rewards/rejected": -1.895032525062561,
      "step": 100
    },
    {
      "epoch": 4.4,
      "grad_norm": 0.15997409765623294,
      "learning_rate": 2.1804183734670277e-07,
      "logits/chosen": -1.4095523357391357,
      "logits/rejected": -1.4386852979660034,
      "logps/chosen": -37.86528778076172,
      "logps/rejected": -281.3670654296875,
      "loss": 0.2041,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.34532371163368225,
      "rewards/margins": 2.3692967891693115,
      "rewards/rejected": -2.023972988128662,
      "step": 110
    },
    {
      "epoch": 4.8,
      "grad_norm": 0.12276969074858735,
      "learning_rate": 2.454718665888589e-08,
      "logits/chosen": -1.541815161705017,
      "logits/rejected": -1.4467297792434692,
      "logps/chosen": -39.434669494628906,
      "logps/rejected": -230.53213500976562,
      "loss": 0.2316,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.4158347249031067,
      "rewards/margins": 2.008779287338257,
      "rewards/rejected": -1.592944622039795,
      "step": 120
    },
    {
      "epoch": 5.0,
      "step": 125,
      "total_flos": 0.0,
      "train_loss": 0.40015009021759035,
      "train_runtime": 1492.5212,
      "train_samples_per_second": 2.68,
      "train_steps_per_second": 0.084
    }
  ],
  "logging_steps": 10,
  "max_steps": 125,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 1000,
  "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": 2,
  "trial_name": null,
  "trial_params": null
}