{
  "best_metric": 1.2733838558197021,
  "best_model_checkpoint": "./outputs/checkpoint-3200",
  "epoch": 2.3315118397085612,
  "eval_steps": 100,
  "global_step": 3200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 2.2671,
      "step": 100
    },
    {
      "epoch": 0.07,
      "eval_loss": 2.1813676357269287,
      "eval_runtime": 549.6637,
      "eval_samples_per_second": 11.414,
      "eval_steps_per_second": 1.428,
      "step": 100
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 2.1547,
      "step": 200
    },
    {
      "epoch": 0.15,
      "eval_loss": 2.131509304046631,
      "eval_runtime": 549.883,
      "eval_samples_per_second": 11.41,
      "eval_steps_per_second": 1.428,
      "step": 200
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 2.1088,
      "step": 300
    },
    {
      "epoch": 0.22,
      "eval_loss": 2.0894575119018555,
      "eval_runtime": 550.3941,
      "eval_samples_per_second": 11.399,
      "eval_steps_per_second": 1.426,
      "step": 300
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 2.0681,
      "step": 400
    },
    {
      "epoch": 0.29,
      "eval_loss": 2.0435657501220703,
      "eval_runtime": 550.6916,
      "eval_samples_per_second": 11.393,
      "eval_steps_per_second": 1.425,
      "step": 400
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 2.0241,
      "step": 500
    },
    {
      "epoch": 0.36,
      "eval_loss": 2.0031208992004395,
      "eval_runtime": 550.5855,
      "eval_samples_per_second": 11.395,
      "eval_steps_per_second": 1.426,
      "step": 500
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 1.9755,
      "step": 600
    },
    {
      "epoch": 0.44,
      "eval_loss": 1.9635978937149048,
      "eval_runtime": 552.4575,
      "eval_samples_per_second": 11.357,
      "eval_steps_per_second": 1.421,
      "step": 600
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 1.9447,
      "step": 700
    },
    {
      "epoch": 0.51,
      "eval_loss": 1.9280308485031128,
      "eval_runtime": 552.7859,
      "eval_samples_per_second": 11.35,
      "eval_steps_per_second": 1.42,
      "step": 700
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 1.9024,
      "step": 800
    },
    {
      "epoch": 0.58,
      "eval_loss": 1.8917897939682007,
      "eval_runtime": 553.0559,
      "eval_samples_per_second": 11.344,
      "eval_steps_per_second": 1.419,
      "step": 800
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002,
      "loss": 1.8626,
      "step": 900
    },
    {
      "epoch": 0.66,
      "eval_loss": 1.8584532737731934,
      "eval_runtime": 554.4522,
      "eval_samples_per_second": 11.316,
      "eval_steps_per_second": 1.416,
      "step": 900
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 1.8397,
      "step": 1000
    },
    {
      "epoch": 0.73,
      "eval_loss": 1.8282623291015625,
      "eval_runtime": 554.0635,
      "eval_samples_per_second": 11.324,
      "eval_steps_per_second": 1.417,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 1.8218,
      "step": 1100
    },
    {
      "epoch": 0.8,
      "eval_loss": 1.7944462299346924,
      "eval_runtime": 554.2451,
      "eval_samples_per_second": 11.32,
      "eval_steps_per_second": 1.416,
      "step": 1100
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 1.7698,
      "step": 1200
    },
    {
      "epoch": 0.87,
      "eval_loss": 1.7612203359603882,
      "eval_runtime": 554.1779,
      "eval_samples_per_second": 11.321,
      "eval_steps_per_second": 1.417,
      "step": 1200
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002,
      "loss": 1.7685,
      "step": 1300
    },
    {
      "epoch": 0.95,
      "eval_loss": 1.730454921722412,
      "eval_runtime": 554.4409,
      "eval_samples_per_second": 11.316,
      "eval_steps_per_second": 1.416,
      "step": 1300
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 1.6896,
      "step": 1400
    },
    {
      "epoch": 1.02,
      "eval_loss": 1.691584587097168,
      "eval_runtime": 554.1975,
      "eval_samples_per_second": 11.321,
      "eval_steps_per_second": 1.416,
      "step": 1400
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002,
      "loss": 1.6213,
      "step": 1500
    },
    {
      "epoch": 1.09,
      "eval_loss": 1.6627237796783447,
      "eval_runtime": 554.2183,
      "eval_samples_per_second": 11.32,
      "eval_steps_per_second": 1.416,
      "step": 1500
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 1.6016,
      "step": 1600
    },
    {
      "epoch": 1.17,
      "eval_loss": 1.6322499513626099,
      "eval_runtime": 554.3251,
      "eval_samples_per_second": 11.318,
      "eval_steps_per_second": 1.416,
      "step": 1600
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002,
      "loss": 1.5939,
      "step": 1700
    },
    {
      "epoch": 1.24,
      "eval_loss": 1.6027921438217163,
      "eval_runtime": 555.8406,
      "eval_samples_per_second": 11.287,
      "eval_steps_per_second": 1.412,
      "step": 1700
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002,
      "loss": 1.5416,
      "step": 1800
    },
    {
      "epoch": 1.31,
      "eval_loss": 1.5798810720443726,
      "eval_runtime": 555.8321,
      "eval_samples_per_second": 11.288,
      "eval_steps_per_second": 1.412,
      "step": 1800
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 1.5149,
      "step": 1900
    },
    {
      "epoch": 1.38,
      "eval_loss": 1.5527715682983398,
      "eval_runtime": 556.317,
      "eval_samples_per_second": 11.278,
      "eval_steps_per_second": 1.411,
      "step": 1900
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0002,
      "loss": 1.5057,
      "step": 2000
    },
    {
      "epoch": 1.46,
      "eval_loss": 1.5266777276992798,
      "eval_runtime": 556.4442,
      "eval_samples_per_second": 11.275,
      "eval_steps_per_second": 1.411,
      "step": 2000
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 1.4851,
      "step": 2100
    },
    {
      "epoch": 1.53,
      "eval_loss": 1.5026806592941284,
      "eval_runtime": 556.1829,
      "eval_samples_per_second": 11.28,
      "eval_steps_per_second": 1.411,
      "step": 2100
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 1.4473,
      "step": 2200
    },
    {
      "epoch": 1.6,
      "eval_loss": 1.4737541675567627,
      "eval_runtime": 555.4267,
      "eval_samples_per_second": 11.296,
      "eval_steps_per_second": 1.413,
      "step": 2200
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002,
      "loss": 1.426,
      "step": 2300
    },
    {
      "epoch": 1.68,
      "eval_loss": 1.4537303447723389,
      "eval_runtime": 555.9884,
      "eval_samples_per_second": 11.284,
      "eval_steps_per_second": 1.412,
      "step": 2300
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0002,
      "loss": 1.4171,
      "step": 2400
    },
    {
      "epoch": 1.75,
      "eval_loss": 1.431217908859253,
      "eval_runtime": 557.2566,
      "eval_samples_per_second": 11.259,
      "eval_steps_per_second": 1.409,
      "step": 2400
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 1.3827,
      "step": 2500
    },
    {
      "epoch": 1.82,
      "eval_loss": 1.4048686027526855,
      "eval_runtime": 555.8745,
      "eval_samples_per_second": 11.287,
      "eval_steps_per_second": 1.412,
      "step": 2500
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0002,
      "loss": 1.3655,
      "step": 2600
    },
    {
      "epoch": 1.89,
      "eval_loss": 1.3847733736038208,
      "eval_runtime": 556.7868,
      "eval_samples_per_second": 11.268,
      "eval_steps_per_second": 1.41,
      "step": 2600
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0002,
      "loss": 1.3487,
      "step": 2700
    },
    {
      "epoch": 1.97,
      "eval_loss": 1.3603800535202026,
      "eval_runtime": 556.2959,
      "eval_samples_per_second": 11.278,
      "eval_steps_per_second": 1.411,
      "step": 2700
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0002,
      "loss": 1.2769,
      "step": 2800
    },
    {
      "epoch": 2.04,
      "eval_loss": 1.3475805521011353,
      "eval_runtime": 556.1265,
      "eval_samples_per_second": 11.282,
      "eval_steps_per_second": 1.412,
      "step": 2800
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0002,
      "loss": 1.226,
      "step": 2900
    },
    {
      "epoch": 2.11,
      "eval_loss": 1.3265427350997925,
      "eval_runtime": 555.0977,
      "eval_samples_per_second": 11.303,
      "eval_steps_per_second": 1.414,
      "step": 2900
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0002,
      "loss": 1.2435,
      "step": 3000
    },
    {
      "epoch": 2.19,
      "eval_loss": 1.3085519075393677,
      "eval_runtime": 554.9705,
      "eval_samples_per_second": 11.305,
      "eval_steps_per_second": 1.414,
      "step": 3000
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0002,
      "loss": 1.2063,
      "step": 3100
    },
    {
      "epoch": 2.26,
      "eval_loss": 1.2864030599594116,
      "eval_runtime": 555.8466,
      "eval_samples_per_second": 11.287,
      "eval_steps_per_second": 1.412,
      "step": 3100
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0002,
      "loss": 1.2078,
      "step": 3200
    },
    {
      "epoch": 2.33,
      "eval_loss": 1.2733838558197021,
      "eval_runtime": 556.0489,
      "eval_samples_per_second": 11.283,
      "eval_steps_per_second": 1.412,
      "step": 3200
    }
  ],
  "logging_steps": 100,
  "max_steps": 4116,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 8.332718824760279e+17,
  "trial_name": null,
  "trial_params": null
}