{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9372888823615804,
  "global_step": 20000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "learning_rate": 1.9510451852939737e-05,
      "loss": 3.3474,
      "step": 500
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9020903705879475e-05,
      "loss": 2.9429,
      "step": 1000
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.853135555881921e-05,
      "loss": 2.8484,
      "step": 1500
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8041807411758948e-05,
      "loss": 2.8794,
      "step": 2000
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7552259264698686e-05,
      "loss": 2.7318,
      "step": 2500
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.706271111763842e-05,
      "loss": 2.7826,
      "step": 3000
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.657316297057816e-05,
      "loss": 2.6663,
      "step": 3500
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6083614823517894e-05,
      "loss": 2.6629,
      "step": 4000
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5594066676457633e-05,
      "loss": 2.8061,
      "step": 4500
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5104518529397367e-05,
      "loss": 2.6693,
      "step": 5000
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4614970382337104e-05,
      "loss": 2.6816,
      "step": 5500
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.412542223527684e-05,
      "loss": 2.6255,
      "step": 6000
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3635874088216576e-05,
      "loss": 2.5631,
      "step": 6500
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.345140218734741,
      "eval_runtime": 23.5405,
      "eval_samples_per_second": 257.089,
      "step": 6809
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3146325941156315e-05,
      "loss": 2.6174,
      "step": 7000
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2656777794096052e-05,
      "loss": 2.6891,
      "step": 7500
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2167229647035787e-05,
      "loss": 2.6243,
      "step": 8000
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1677681499975524e-05,
      "loss": 2.6427,
      "step": 8500
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.118813335291526e-05,
      "loss": 2.5105,
      "step": 9000
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0698585205854997e-05,
      "loss": 2.4671,
      "step": 9500
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0209037058794733e-05,
      "loss": 2.5177,
      "step": 10000
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.71948891173447e-06,
      "loss": 2.5607,
      "step": 10500
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.229940764674206e-06,
      "loss": 2.596,
      "step": 11000
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.740392617613943e-06,
      "loss": 2.5601,
      "step": 11500
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.25084447055368e-06,
      "loss": 2.5039,
      "step": 12000
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.761296323493416e-06,
      "loss": 2.4822,
      "step": 12500
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.271748176433154e-06,
      "loss": 2.6403,
      "step": 13000
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.782200029372889e-06,
      "loss": 2.5482,
      "step": 13500
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.317995071411133,
      "eval_runtime": 23.5909,
      "eval_samples_per_second": 256.54,
      "step": 13618
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.292651882312626e-06,
      "loss": 2.4993,
      "step": 14000
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.8031037352523625e-06,
      "loss": 2.4475,
      "step": 14500
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.313555588192098e-06,
      "loss": 2.5505,
      "step": 15000
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.824007441131836e-06,
      "loss": 2.5814,
      "step": 15500
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.334459294071572e-06,
      "loss": 2.5055,
      "step": 16000
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.844911147011309e-06,
      "loss": 2.4834,
      "step": 16500
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3553629999510454e-06,
      "loss": 2.4866,
      "step": 17000
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.865814852890782e-06,
      "loss": 2.5307,
      "step": 17500
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.3762667058305186e-06,
      "loss": 2.5142,
      "step": 18000
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8867185587702552e-06,
      "loss": 2.5078,
      "step": 18500
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.3971704117099918e-06,
      "loss": 2.5071,
      "step": 19000
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.076222646497284e-07,
      "loss": 2.5398,
      "step": 19500
    },
    {
      "epoch": 2.94,
      "learning_rate": 4.1807411758946493e-07,
      "loss": 2.5442,
      "step": 20000
    }
  ],
  "max_steps": 20427,
  "num_train_epochs": 3,
  "total_flos": 2163835150460172.0,
  "trial_name": null,
  "trial_params": null
}