{
  "best_metric": 0.2948146462440491,
  "best_model_checkpoint": "ckpt/llama3_8b_other/object_counting_no_sys/checkpoint-400",
  "epoch": 6.0,
  "eval_steps": 200,
  "global_step": 510,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11764705882352941,
      "grad_norm": 23.17191505432129,
      "learning_rate": 2.5e-05,
      "loss": 7.0189,
      "step": 10
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 8.94247055053711,
      "learning_rate": 5e-05,
      "loss": 3.2984,
      "step": 20
    },
    {
      "epoch": 0.35294117647058826,
      "grad_norm": 9.85525894165039,
      "learning_rate": 4.994863481875841e-05,
      "loss": 0.7046,
      "step": 30
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 5.321958065032959,
      "learning_rate": 4.979475034558115e-05,
      "loss": 0.5767,
      "step": 40
    },
    {
      "epoch": 0.5882352941176471,
      "grad_norm": 8.481945037841797,
      "learning_rate": 4.9538978924776634e-05,
      "loss": 0.5266,
      "step": 50
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 18.951034545898438,
      "learning_rate": 4.9182371575975736e-05,
      "loss": 0.5439,
      "step": 60
    },
    {
      "epoch": 0.8235294117647058,
      "grad_norm": 8.39922046661377,
      "learning_rate": 4.8726393675266716e-05,
      "loss": 0.4045,
      "step": 70
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 7.662684917449951,
      "learning_rate": 4.817291893365055e-05,
      "loss": 0.3646,
      "step": 80
    },
    {
      "epoch": 1.0588235294117647,
      "grad_norm": 5.594130992889404,
      "learning_rate": 4.752422169756048e-05,
      "loss": 0.4374,
      "step": 90
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 14.85095500946045,
      "learning_rate": 4.678296760308474e-05,
      "loss": 0.3521,
      "step": 100
    },
    {
      "epoch": 1.2941176470588236,
      "grad_norm": 7.1303606033325195,
      "learning_rate": 4.595220262229601e-05,
      "loss": 0.3584,
      "step": 110
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 9.753626823425293,
      "learning_rate": 4.503534054669892e-05,
      "loss": 0.4017,
      "step": 120
    },
    {
      "epoch": 1.5294117647058822,
      "grad_norm": 8.411559104919434,
      "learning_rate": 4.4036148959228365e-05,
      "loss": 0.4192,
      "step": 130
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 10.687651634216309,
      "learning_rate": 4.2958733752443195e-05,
      "loss": 0.3832,
      "step": 140
    },
    {
      "epoch": 1.7647058823529411,
      "grad_norm": 15.22123908996582,
      "learning_rate": 4.180752225653292e-05,
      "loss": 0.2871,
      "step": 150
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 7.608785629272461,
      "learning_rate": 4.058724504646834e-05,
      "loss": 0.2344,
      "step": 160
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.1389312744140625,
      "learning_rate": 3.9302916503054246e-05,
      "loss": 0.2686,
      "step": 170
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 25.7930908203125,
      "learning_rate": 3.7959814207763135e-05,
      "loss": 0.3129,
      "step": 180
    },
    {
      "epoch": 2.235294117647059,
      "grad_norm": 4.136214733123779,
      "learning_rate": 3.656345725602089e-05,
      "loss": 0.2146,
      "step": 190
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 3.397460699081421,
      "learning_rate": 3.5119583578059846e-05,
      "loss": 0.1766,
      "step": 200
    },
    {
      "epoch": 2.3529411764705883,
      "eval_loss": 0.4584803879261017,
      "eval_runtime": 0.7658,
      "eval_samples_per_second": 156.69,
      "eval_steps_per_second": 19.586,
      "step": 200
    },
    {
      "epoch": 2.4705882352941178,
      "grad_norm": 15.182618141174316,
      "learning_rate": 3.363412636053269e-05,
      "loss": 0.1984,
      "step": 210
    },
    {
      "epoch": 2.588235294117647,
      "grad_norm": 7.012846946716309,
      "learning_rate": 3.211318966577581e-05,
      "loss": 0.1822,
      "step": 220
    },
    {
      "epoch": 2.7058823529411766,
      "grad_norm": 10.27369499206543,
      "learning_rate": 3.056302334890786e-05,
      "loss": 0.2768,
      "step": 230
    },
    {
      "epoch": 2.8235294117647056,
      "grad_norm": 7.723974704742432,
      "learning_rate": 2.8989997375834482e-05,
      "loss": 0.1146,
      "step": 240
    },
    {
      "epoch": 2.9411764705882355,
      "grad_norm": 16.098949432373047,
      "learning_rate": 2.7400575647692046e-05,
      "loss": 0.2341,
      "step": 250
    },
    {
      "epoch": 3.0588235294117645,
      "grad_norm": 3.3051462173461914,
      "learning_rate": 2.5801289439291388e-05,
      "loss": 0.2811,
      "step": 260
    },
    {
      "epoch": 3.176470588235294,
      "grad_norm": 14.433158874511719,
      "learning_rate": 2.419871056070862e-05,
      "loss": 0.1915,
      "step": 270
    },
    {
      "epoch": 3.2941176470588234,
      "grad_norm": 3.7534849643707275,
      "learning_rate": 2.2599424352307957e-05,
      "loss": 0.1292,
      "step": 280
    },
    {
      "epoch": 3.411764705882353,
      "grad_norm": 15.175636291503906,
      "learning_rate": 2.1010002624165527e-05,
      "loss": 0.1015,
      "step": 290
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 11.786360740661621,
      "learning_rate": 1.9436976651092144e-05,
      "loss": 0.1129,
      "step": 300
    },
    {
      "epoch": 3.6470588235294117,
      "grad_norm": 13.687607765197754,
      "learning_rate": 1.7886810334224192e-05,
      "loss": 0.1072,
      "step": 310
    },
    {
      "epoch": 3.764705882352941,
      "grad_norm": 6.054112434387207,
      "learning_rate": 1.6365873639467315e-05,
      "loss": 0.1333,
      "step": 320
    },
    {
      "epoch": 3.8823529411764706,
      "grad_norm": 12.263951301574707,
      "learning_rate": 1.4880416421940155e-05,
      "loss": 0.1509,
      "step": 330
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.6237934827804565,
      "learning_rate": 1.3436542743979125e-05,
      "loss": 0.0754,
      "step": 340
    },
    {
      "epoch": 4.117647058823529,
      "grad_norm": 2.395211935043335,
      "learning_rate": 1.2040185792236874e-05,
      "loss": 0.0869,
      "step": 350
    },
    {
      "epoch": 4.235294117647059,
      "grad_norm": 7.990754127502441,
      "learning_rate": 1.0697083496945765e-05,
      "loss": 0.1263,
      "step": 360
    },
    {
      "epoch": 4.352941176470588,
      "grad_norm": 5.7818779945373535,
      "learning_rate": 9.412754953531663e-06,
      "loss": 0.0479,
      "step": 370
    },
    {
      "epoch": 4.470588235294118,
      "grad_norm": 8.458385467529297,
      "learning_rate": 8.192477743467078e-06,
      "loss": 0.0401,
      "step": 380
    },
    {
      "epoch": 4.588235294117647,
      "grad_norm": 4.223268032073975,
      "learning_rate": 7.041266247556813e-06,
      "loss": 0.0774,
      "step": 390
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 2.8907740116119385,
      "learning_rate": 5.9638510407716394e-06,
      "loss": 0.0726,
      "step": 400
    },
    {
      "epoch": 4.705882352941177,
      "eval_loss": 0.2948146462440491,
      "eval_runtime": 0.7657,
      "eval_samples_per_second": 156.723,
      "eval_steps_per_second": 19.59,
      "step": 400
    },
    {
      "epoch": 4.823529411764706,
      "grad_norm": 7.116231441497803,
      "learning_rate": 4.9646594533010875e-06,
      "loss": 0.0701,
      "step": 410
    },
    {
      "epoch": 4.9411764705882355,
      "grad_norm": 15.119770050048828,
      "learning_rate": 4.047797377703985e-06,
      "loss": 0.0393,
      "step": 420
    },
    {
      "epoch": 5.0588235294117645,
      "grad_norm": 11.085099220275879,
      "learning_rate": 3.217032396915265e-06,
      "loss": 0.0839,
      "step": 430
    },
    {
      "epoch": 5.176470588235294,
      "grad_norm": 4.78342342376709,
      "learning_rate": 2.475778302439524e-06,
      "loss": 0.0348,
      "step": 440
    },
    {
      "epoch": 5.294117647058823,
      "grad_norm": 0.49753427505493164,
      "learning_rate": 1.827081066349459e-06,
      "loss": 0.0429,
      "step": 450
    },
    {
      "epoch": 5.411764705882353,
      "grad_norm": 4.4141693115234375,
      "learning_rate": 1.273606324733284e-06,
      "loss": 0.0493,
      "step": 460
    },
    {
      "epoch": 5.529411764705882,
      "grad_norm": 5.7931413650512695,
      "learning_rate": 8.176284240242638e-07,
      "loss": 0.0158,
      "step": 470
    },
    {
      "epoch": 5.647058823529412,
      "grad_norm": 6.445806980133057,
      "learning_rate": 4.6102107522336403e-07,
      "loss": 0.026,
      "step": 480
    },
    {
      "epoch": 5.764705882352941,
      "grad_norm": 0.24533455073833466,
      "learning_rate": 2.052496544188487e-07,
      "loss": 0.0267,
      "step": 490
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 13.630054473876953,
      "learning_rate": 5.136518124159162e-08,
      "loss": 0.0317,
      "step": 500
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.213211178779602,
      "learning_rate": 0.0,
      "loss": 0.0259,
      "step": 510
    },
    {
      "epoch": 6.0,
      "step": 510,
      "total_flos": 1.0415358657691648e+16,
      "train_loss": 0.4013917989882768,
      "train_runtime": 114.995,
      "train_samples_per_second": 35.48,
      "train_steps_per_second": 4.435
    }
  ],
  "logging_steps": 10,
  "max_steps": 510,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 200,
  "total_flos": 1.0415358657691648e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}