{
  "best_metric": 0.0060875071212649345,
  "best_model_checkpoint": "./Zephyr/28-03-24-Weni-WeniGPT-QA-Zephyr-7B-4.0.2-KTO_WeniGPT Experiment using KTO trainer with no collator, Zephyr model and random system prompt.-2_max_steps-786_batch_32_2024-03-28_ppid_9/checkpoint-500",
  "epoch": 3.8058991436726926,
  "eval_steps": 50,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15,
      "grad_norm": 3.973484516143799,
      "kl": 0.07447954267263412,
      "learning_rate": 0.00015833333333333332,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 1.3654,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 20
    },
    {
      "epoch": 0.3,
      "grad_norm": 2.1150100231170654,
      "kl": 0.24954533576965332,
      "learning_rate": 0.00019606299212598428,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.1911,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 40
    },
    {
      "epoch": 0.38,
      "eval_kl": 0.0,
      "eval_logps/chosen": -157.132080078125,
      "eval_logps/rejected": -431.1063232421875,
      "eval_loss": 0.019795970991253853,
      "eval_rewards/chosen": 5.025023460388184,
      "eval_rewards/margins": 28.0367374420166,
      "eval_rewards/rejected": -23.0117130279541,
      "eval_runtime": 209.568,
      "eval_samples_per_second": 2.252,
      "eval_steps_per_second": 0.563,
      "step": 50
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.8193832039833069,
      "kl": 0.0,
      "learning_rate": 0.00019081364829396326,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.1212,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 60
    },
    {
      "epoch": 0.61,
      "grad_norm": 1.1438875198364258,
      "kl": 0.0,
      "learning_rate": 0.00018556430446194227,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.0613,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 80
    },
    {
      "epoch": 0.76,
      "grad_norm": 1.5549229383468628,
      "kl": 0.30086809396743774,
      "learning_rate": 0.00018031496062992125,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.0525,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 100
    },
    {
      "epoch": 0.76,
      "eval_kl": 0.0,
      "eval_logps/chosen": -151.0825958251953,
      "eval_logps/rejected": -502.1325378417969,
      "eval_loss": 0.015516282990574837,
      "eval_rewards/chosen": 5.629973411560059,
      "eval_rewards/margins": 35.744300842285156,
      "eval_rewards/rejected": -30.11433219909668,
      "eval_runtime": 209.5342,
      "eval_samples_per_second": 2.253,
      "eval_steps_per_second": 0.563,
      "step": 100
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.16322359442710876,
      "kl": 0.0,
      "learning_rate": 0.0001750656167979003,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.037,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 120
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.8122725486755371,
      "kl": 0.0,
      "learning_rate": 0.00016981627296587927,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.0684,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 140
    },
    {
      "epoch": 1.14,
      "eval_kl": 0.0,
      "eval_logps/chosen": -145.42347717285156,
      "eval_logps/rejected": -530.3358154296875,
      "eval_loss": 0.010358058847486973,
      "eval_rewards/chosen": 6.19588565826416,
      "eval_rewards/margins": 39.130550384521484,
      "eval_rewards/rejected": -32.934661865234375,
      "eval_runtime": 209.5283,
      "eval_samples_per_second": 2.253,
      "eval_steps_per_second": 0.563,
      "step": 150
    },
    {
      "epoch": 1.22,
      "grad_norm": 0.11464398354291916,
      "kl": 0.0,
      "learning_rate": 0.00016456692913385828,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.0167,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 160
    },
    {
      "epoch": 1.37,
      "grad_norm": 5.260560989379883,
      "kl": 0.0,
      "learning_rate": 0.00015931758530183726,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.131,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 180
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.15409517288208008,
      "kl": 0.0,
      "learning_rate": 0.0001540682414698163,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.0253,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 200
    },
    {
      "epoch": 1.52,
      "eval_kl": 0.0,
      "eval_logps/chosen": -147.8525390625,
      "eval_logps/rejected": -489.1932678222656,
      "eval_loss": 0.012590945698320866,
      "eval_rewards/chosen": 5.952979564666748,
      "eval_rewards/margins": 34.77338790893555,
      "eval_rewards/rejected": -28.82040786743164,
      "eval_runtime": 209.5044,
      "eval_samples_per_second": 2.253,
      "eval_steps_per_second": 0.563,
      "step": 200
    },
    {
      "epoch": 1.67,
      "grad_norm": 0.025852208957076073,
      "kl": 0.0,
      "learning_rate": 0.00014908136482939634,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.0101,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 220
    },
    {
      "epoch": 1.83,
      "grad_norm": 0.013888155110180378,
      "kl": 0.1387307345867157,
      "learning_rate": 0.00014383202099737535,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.0177,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 240
    },
    {
      "epoch": 1.9,
      "eval_kl": 0.0,
      "eval_logps/chosen": -142.92942810058594,
      "eval_logps/rejected": -740.499267578125,
      "eval_loss": 0.01631937175989151,
      "eval_rewards/chosen": 6.445290565490723,
      "eval_rewards/margins": 60.39630126953125,
      "eval_rewards/rejected": -53.951011657714844,
      "eval_runtime": 209.4415,
      "eval_samples_per_second": 2.254,
      "eval_steps_per_second": 0.563,
      "step": 250
    },
    {
      "epoch": 1.98,
      "grad_norm": 0.02793472446501255,
      "kl": 0.0,
      "learning_rate": 0.00013858267716535433,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.0112,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 260
    },
    {
      "epoch": 2.13,
      "grad_norm": 0.0313153937458992,
      "kl": 0.0,
      "learning_rate": 0.00013333333333333334,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.0067,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 280
    },
    {
      "epoch": 2.28,
      "grad_norm": 0.030554356053471565,
      "kl": 0.0,
      "learning_rate": 0.00012808398950131235,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.0126,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 300
    },
    {
      "epoch": 2.28,
      "eval_kl": 0.0,
      "eval_logps/chosen": -145.38067626953125,
      "eval_logps/rejected": -851.4735717773438,
      "eval_loss": 0.020638220012187958,
      "eval_rewards/chosen": 6.200164794921875,
      "eval_rewards/margins": 71.24861145019531,
      "eval_rewards/rejected": -65.04844665527344,
      "eval_runtime": 209.5591,
      "eval_samples_per_second": 2.252,
      "eval_steps_per_second": 0.563,
      "step": 300
    },
    {
      "epoch": 2.44,
      "grad_norm": 0.01577291637659073,
      "kl": 0.0,
      "learning_rate": 0.00012283464566929136,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.0134,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 320
    },
    {
      "epoch": 2.59,
      "grad_norm": 0.017728939652442932,
      "kl": 0.0,
      "learning_rate": 0.00011758530183727034,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.007,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 340
    },
    {
      "epoch": 2.66,
      "eval_kl": 0.0,
      "eval_logps/chosen": -140.51016235351562,
      "eval_logps/rejected": -616.1578369140625,
      "eval_loss": 0.006825109478086233,
      "eval_rewards/chosen": 6.687215328216553,
      "eval_rewards/margins": 48.20407485961914,
      "eval_rewards/rejected": -41.51686477661133,
      "eval_runtime": 209.4969,
      "eval_samples_per_second": 2.253,
      "eval_steps_per_second": 0.563,
      "step": 350
    },
    {
      "epoch": 2.74,
      "grad_norm": 0.025538522750139236,
      "kl": 0.2244974821805954,
      "learning_rate": 0.00011233595800524934,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.0034,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 360
    },
    {
      "epoch": 2.89,
      "grad_norm": 0.009798596613109112,
      "kl": 0.0,
      "learning_rate": 0.00010708661417322836,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.0081,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 380
    },
    {
      "epoch": 3.04,
      "grad_norm": 0.021639486774802208,
      "kl": 0.0,
      "learning_rate": 0.00010183727034120735,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.0085,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 400
    },
    {
      "epoch": 3.04,
      "eval_kl": 0.0,
      "eval_logps/chosen": -139.7056884765625,
      "eval_logps/rejected": -727.2117309570312,
      "eval_loss": 0.007565339561551809,
      "eval_rewards/chosen": 6.767664432525635,
      "eval_rewards/margins": 59.389923095703125,
      "eval_rewards/rejected": -52.62226104736328,
      "eval_runtime": 209.4194,
      "eval_samples_per_second": 2.254,
      "eval_steps_per_second": 0.563,
      "step": 400
    },
    {
      "epoch": 3.2,
      "grad_norm": 0.2425842434167862,
      "kl": 0.0,
      "learning_rate": 9.658792650918635e-05,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.0029,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 420
    },
    {
      "epoch": 3.35,
      "grad_norm": 0.01065791118890047,
      "kl": 0.0,
      "learning_rate": 9.133858267716536e-05,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.0057,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 440
    },
    {
      "epoch": 3.43,
      "eval_kl": 0.0,
      "eval_logps/chosen": -141.96299743652344,
      "eval_logps/rejected": -582.8775024414062,
      "eval_loss": 0.005958274472504854,
      "eval_rewards/chosen": 6.541934967041016,
      "eval_rewards/margins": 44.73076629638672,
      "eval_rewards/rejected": -38.18882751464844,
      "eval_runtime": 209.4524,
      "eval_samples_per_second": 2.253,
      "eval_steps_per_second": 0.563,
      "step": 450
    },
    {
      "epoch": 3.5,
      "grad_norm": 0.015213225036859512,
      "kl": 0.0,
      "learning_rate": 8.608923884514435e-05,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.008,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 460
    },
    {
      "epoch": 3.65,
      "grad_norm": 0.016072571277618408,
      "kl": 0.0,
      "learning_rate": 8.083989501312336e-05,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.0119,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 480
    },
    {
      "epoch": 3.81,
      "grad_norm": 0.007641986012458801,
      "kl": 0.16172848641872406,
      "learning_rate": 7.559055118110236e-05,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 0.0029,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 500
    },
    {
      "epoch": 3.81,
      "eval_kl": 0.0,
      "eval_logps/chosen": -140.1928253173828,
      "eval_logps/rejected": -571.4075927734375,
      "eval_loss": 0.0060875071212649345,
      "eval_rewards/chosen": 6.718951225280762,
      "eval_rewards/margins": 43.76079177856445,
      "eval_rewards/rejected": -37.041839599609375,
      "eval_runtime": 209.4408,
      "eval_samples_per_second": 2.254,
      "eval_steps_per_second": 0.563,
      "step": 500
    }
  ],
  "logging_steps": 20,
  "max_steps": 786,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}