{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.9873015873015873,
  "eval_steps": 500,
  "global_step": 628,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06,
      "learning_rate": 7.936507936507936e-08,
      "loss": 6.6074,
      "step": 10
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.5873015873015872e-07,
      "loss": 4.4017,
      "step": 20
    },
    {
      "epoch": 0.19,
      "learning_rate": 2.3809523809523806e-07,
      "loss": 1.5627,
      "step": 30
    },
    {
      "epoch": 0.25,
      "learning_rate": 3.1746031746031743e-07,
      "loss": 0.5796,
      "step": 40
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.968253968253968e-07,
      "loss": 0.2248,
      "step": 50
    },
    {
      "epoch": 0.38,
      "learning_rate": 4.761904761904761e-07,
      "loss": 0.1782,
      "step": 60
    },
    {
      "epoch": 0.44,
      "learning_rate": 4.998106548810311e-07,
      "loss": 0.1544,
      "step": 70
    },
    {
      "epoch": 0.51,
      "learning_rate": 4.988839406031596e-07,
      "loss": 0.119,
      "step": 80
    },
    {
      "epoch": 0.57,
      "learning_rate": 4.971879403278432e-07,
      "loss": 0.0793,
      "step": 90
    },
    {
      "epoch": 0.63,
      "learning_rate": 4.947278962947386e-07,
      "loss": 0.0606,
      "step": 100
    },
    {
      "epoch": 0.7,
      "learning_rate": 4.915114123589732e-07,
      "loss": 0.0565,
      "step": 110
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.875484304880629e-07,
      "loss": 0.0616,
      "step": 120
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.828512000318616e-07,
      "loss": 0.0552,
      "step": 130
    },
    {
      "epoch": 0.89,
      "learning_rate": 4.774342398605221e-07,
      "loss": 0.0562,
      "step": 140
    },
    {
      "epoch": 0.95,
      "learning_rate": 4.713142934875005e-07,
      "loss": 0.0536,
      "step": 150
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.056731484830379486,
      "eval_runtime": 22.0428,
      "eval_samples_per_second": 9.3,
      "eval_steps_per_second": 0.817,
      "step": 157
    },
    {
      "epoch": 1.02,
      "learning_rate": 4.64510277316316e-07,
      "loss": 0.0586,
      "step": 160
    },
    {
      "epoch": 1.08,
      "learning_rate": 4.570432221710314e-07,
      "loss": 0.0494,
      "step": 170
    },
    {
      "epoch": 1.14,
      "learning_rate": 4.4893620829118124e-07,
      "loss": 0.0558,
      "step": 180
    },
    {
      "epoch": 1.21,
      "learning_rate": 4.40214293992074e-07,
      "loss": 0.0514,
      "step": 190
    },
    {
      "epoch": 1.27,
      "learning_rate": 4.3090443821097566e-07,
      "loss": 0.0481,
      "step": 200
    },
    {
      "epoch": 1.33,
      "learning_rate": 4.210354171785795e-07,
      "loss": 0.049,
      "step": 210
    },
    {
      "epoch": 1.4,
      "learning_rate": 4.1063773547332584e-07,
      "loss": 0.0487,
      "step": 220
    },
    {
      "epoch": 1.46,
      "learning_rate": 3.997435317334988e-07,
      "loss": 0.0487,
      "step": 230
    },
    {
      "epoch": 1.52,
      "learning_rate": 3.8838647931853684e-07,
      "loss": 0.0434,
      "step": 240
    },
    {
      "epoch": 1.59,
      "learning_rate": 3.7660168222660824e-07,
      "loss": 0.0464,
      "step": 250
    },
    {
      "epoch": 1.65,
      "learning_rate": 3.6442556659016475e-07,
      "loss": 0.0443,
      "step": 260
    },
    {
      "epoch": 1.71,
      "learning_rate": 3.5189576808485404e-07,
      "loss": 0.0435,
      "step": 270
    },
    {
      "epoch": 1.78,
      "learning_rate": 3.390510155998023e-07,
      "loss": 0.0405,
      "step": 280
    },
    {
      "epoch": 1.84,
      "learning_rate": 3.2593101152883795e-07,
      "loss": 0.0469,
      "step": 290
    },
    {
      "epoch": 1.9,
      "learning_rate": 3.125763090526674e-07,
      "loss": 0.0457,
      "step": 300
    },
    {
      "epoch": 1.97,
      "learning_rate": 2.9902818679131775e-07,
      "loss": 0.044,
      "step": 310
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.045664019882678986,
      "eval_runtime": 22.0358,
      "eval_samples_per_second": 9.303,
      "eval_steps_per_second": 0.817,
      "step": 315
    },
    {
      "epoch": 2.03,
      "learning_rate": 2.8532852121428733e-07,
      "loss": 0.0371,
      "step": 320
    },
    {
      "epoch": 2.1,
      "learning_rate": 2.715196572027789e-07,
      "loss": 0.0345,
      "step": 330
    },
    {
      "epoch": 2.16,
      "learning_rate": 2.5764427716409815e-07,
      "loss": 0.041,
      "step": 340
    },
    {
      "epoch": 2.22,
      "learning_rate": 2.4374526910277886e-07,
      "loss": 0.0379,
      "step": 350
    },
    {
      "epoch": 2.29,
      "learning_rate": 2.2986559405621886e-07,
      "loss": 0.0356,
      "step": 360
    },
    {
      "epoch": 2.35,
      "learning_rate": 2.160481533045751e-07,
      "loss": 0.0333,
      "step": 370
    },
    {
      "epoch": 2.41,
      "learning_rate": 2.0233565576536564e-07,
      "loss": 0.0359,
      "step": 380
    },
    {
      "epoch": 2.48,
      "learning_rate": 1.887704859826528e-07,
      "loss": 0.035,
      "step": 390
    },
    {
      "epoch": 2.54,
      "learning_rate": 1.7539457311884675e-07,
      "loss": 0.0322,
      "step": 400
    },
    {
      "epoch": 2.6,
      "learning_rate": 1.6224926135406693e-07,
      "loss": 0.0385,
      "step": 410
    },
    {
      "epoch": 2.67,
      "learning_rate": 1.4937518209365108e-07,
      "loss": 0.0363,
      "step": 420
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.3681212837880977e-07,
      "loss": 0.0311,
      "step": 430
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.2459893188861613e-07,
      "loss": 0.0405,
      "step": 440
    },
    {
      "epoch": 2.86,
      "learning_rate": 1.1277334291351145e-07,
      "loss": 0.0363,
      "step": 450
    },
    {
      "epoch": 2.92,
      "learning_rate": 1.0137191367132078e-07,
      "loss": 0.0357,
      "step": 460
    },
    {
      "epoch": 2.98,
      "learning_rate": 9.042988532644249e-08,
      "loss": 0.0433,
      "step": 470
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.043395839631557465,
      "eval_runtime": 22.033,
      "eval_samples_per_second": 9.304,
      "eval_steps_per_second": 0.817,
      "step": 472
    },
    {
      "epoch": 3.05,
      "learning_rate": 7.998107906142839e-08,
      "loss": 0.0326,
      "step": 480
    },
    {
      "epoch": 3.11,
      "learning_rate": 7.005779153764682e-08,
      "loss": 0.0327,
      "step": 490
    },
    {
      "epoch": 3.17,
      "learning_rate": 6.069069506815325e-08,
      "loss": 0.0338,
      "step": 500
    },
    {
      "epoch": 3.24,
      "learning_rate": 5.190874281132851e-08,
      "loss": 0.0277,
      "step": 510
    },
    {
      "epoch": 3.3,
      "learning_rate": 4.373907927832513e-08,
      "loss": 0.0266,
      "step": 520
    },
    {
      "epoch": 3.37,
      "learning_rate": 3.620695643093924e-08,
      "loss": 0.032,
      "step": 530
    },
    {
      "epoch": 3.43,
      "learning_rate": 2.9335655629243645e-08,
      "loss": 0.0312,
      "step": 540
    },
    {
      "epoch": 3.49,
      "learning_rate": 2.31464156702382e-08,
      "loss": 0.0324,
      "step": 550
    },
    {
      "epoch": 3.56,
      "learning_rate": 1.7658367139945228e-08,
      "loss": 0.0302,
      "step": 560
    },
    {
      "epoch": 3.62,
      "learning_rate": 1.2888473281864597e-08,
      "loss": 0.0296,
      "step": 570
    },
    {
      "epoch": 3.68,
      "learning_rate": 8.851477564560061e-09,
      "loss": 0.0251,
      "step": 580
    },
    {
      "epoch": 3.75,
      "learning_rate": 5.559858110443016e-09,
      "loss": 0.0299,
      "step": 590
    },
    {
      "epoch": 3.81,
      "learning_rate": 3.023789126611137e-09,
      "loss": 0.0247,
      "step": 600
    },
    {
      "epoch": 3.87,
      "learning_rate": 1.2511094569571668e-09,
      "loss": 0.0296,
      "step": 610
    },
    {
      "epoch": 3.94,
      "learning_rate": 2.4729835275189016e-10,
      "loss": 0.0346,
      "step": 620
    },
    {
      "epoch": 3.99,
      "eval_loss": 0.045722462236881256,
      "eval_runtime": 22.0525,
      "eval_samples_per_second": 9.296,
      "eval_steps_per_second": 0.816,
      "step": 628
    },
    {
      "epoch": 3.99,
      "step": 628,
      "total_flos": 72123112488960.0,
      "train_loss": 0.25595629749119664,
      "train_runtime": 6034.9176,
      "train_samples_per_second": 2.503,
      "train_steps_per_second": 0.104
    }
  ],
  "logging_steps": 10,
  "max_steps": 628,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 72123112488960.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}