{
  "best_metric": 1.1207506656646729,
  "best_model_checkpoint": "./outputs/checkpoint-4100",
  "epoch": 2.987249544626594,
  "eval_steps": 100,
  "global_step": 4100,
  "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": 2.2669,
      "step": 100
    },
    {
      "epoch": 0.07,
      "eval_loss": 2.180582046508789,
      "eval_runtime": 550.6875,
      "eval_samples_per_second": 11.393,
      "eval_steps_per_second": 1.425,
      "step": 100
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 2.1546,
      "step": 200
    },
    {
      "epoch": 0.15,
      "eval_loss": 2.131606101989746,
      "eval_runtime": 550.6761,
      "eval_samples_per_second": 11.393,
      "eval_steps_per_second": 1.426,
      "step": 200
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 2.1084,
      "step": 300
    },
    {
      "epoch": 0.22,
      "eval_loss": 2.089244842529297,
      "eval_runtime": 551.3766,
      "eval_samples_per_second": 11.379,
      "eval_steps_per_second": 1.424,
      "step": 300
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 2.068,
      "step": 400
    },
    {
      "epoch": 0.29,
      "eval_loss": 2.043504476547241,
      "eval_runtime": 551.782,
      "eval_samples_per_second": 11.37,
      "eval_steps_per_second": 1.423,
      "step": 400
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 2.0238,
      "step": 500
    },
    {
      "epoch": 0.36,
      "eval_loss": 2.0026352405548096,
      "eval_runtime": 552.6008,
      "eval_samples_per_second": 11.354,
      "eval_steps_per_second": 1.421,
      "step": 500
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 1.9746,
      "step": 600
    },
    {
      "epoch": 0.44,
      "eval_loss": 1.9627352952957153,
      "eval_runtime": 553.4908,
      "eval_samples_per_second": 11.335,
      "eval_steps_per_second": 1.418,
      "step": 600
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 1.9436,
      "step": 700
    },
    {
      "epoch": 0.51,
      "eval_loss": 1.9262831211090088,
      "eval_runtime": 553.1509,
      "eval_samples_per_second": 11.342,
      "eval_steps_per_second": 1.419,
      "step": 700
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 1.9026,
      "step": 800
    },
    {
      "epoch": 0.58,
      "eval_loss": 1.8915894031524658,
      "eval_runtime": 553.4095,
      "eval_samples_per_second": 11.337,
      "eval_steps_per_second": 1.418,
      "step": 800
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002,
      "loss": 1.8633,
      "step": 900
    },
    {
      "epoch": 0.66,
      "eval_loss": 1.8577481508255005,
      "eval_runtime": 553.9449,
      "eval_samples_per_second": 11.326,
      "eval_steps_per_second": 1.417,
      "step": 900
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 1.8404,
      "step": 1000
    },
    {
      "epoch": 0.73,
      "eval_loss": 1.8280558586120605,
      "eval_runtime": 554.6062,
      "eval_samples_per_second": 11.313,
      "eval_steps_per_second": 1.415,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 1.8207,
      "step": 1100
    },
    {
      "epoch": 0.8,
      "eval_loss": 1.7947672605514526,
      "eval_runtime": 555.1319,
      "eval_samples_per_second": 11.302,
      "eval_steps_per_second": 1.414,
      "step": 1100
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 1.769,
      "step": 1200
    },
    {
      "epoch": 0.87,
      "eval_loss": 1.761672019958496,
      "eval_runtime": 555.3259,
      "eval_samples_per_second": 11.298,
      "eval_steps_per_second": 1.414,
      "step": 1200
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002,
      "loss": 1.7687,
      "step": 1300
    },
    {
      "epoch": 0.95,
      "eval_loss": 1.7306458950042725,
      "eval_runtime": 554.8996,
      "eval_samples_per_second": 11.307,
      "eval_steps_per_second": 1.415,
      "step": 1300
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 1.6906,
      "step": 1400
    },
    {
      "epoch": 1.02,
      "eval_loss": 1.6910051107406616,
      "eval_runtime": 555.5658,
      "eval_samples_per_second": 11.293,
      "eval_steps_per_second": 1.413,
      "step": 1400
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002,
      "loss": 1.6215,
      "step": 1500
    },
    {
      "epoch": 1.09,
      "eval_loss": 1.6626144647598267,
      "eval_runtime": 555.8884,
      "eval_samples_per_second": 11.286,
      "eval_steps_per_second": 1.412,
      "step": 1500
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 1.601,
      "step": 1600
    },
    {
      "epoch": 1.17,
      "eval_loss": 1.6326826810836792,
      "eval_runtime": 555.8386,
      "eval_samples_per_second": 11.287,
      "eval_steps_per_second": 1.412,
      "step": 1600
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002,
      "loss": 1.595,
      "step": 1700
    },
    {
      "epoch": 1.24,
      "eval_loss": 1.6032005548477173,
      "eval_runtime": 555.9827,
      "eval_samples_per_second": 11.285,
      "eval_steps_per_second": 1.412,
      "step": 1700
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002,
      "loss": 1.5417,
      "step": 1800
    },
    {
      "epoch": 1.31,
      "eval_loss": 1.581026315689087,
      "eval_runtime": 556.3673,
      "eval_samples_per_second": 11.277,
      "eval_steps_per_second": 1.411,
      "step": 1800
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 1.5177,
      "step": 1900
    },
    {
      "epoch": 1.38,
      "eval_loss": 1.5540947914123535,
      "eval_runtime": 556.1818,
      "eval_samples_per_second": 11.28,
      "eval_steps_per_second": 1.411,
      "step": 1900
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0002,
      "loss": 1.5071,
      "step": 2000
    },
    {
      "epoch": 1.46,
      "eval_loss": 1.529414415359497,
      "eval_runtime": 556.1478,
      "eval_samples_per_second": 11.281,
      "eval_steps_per_second": 1.411,
      "step": 2000
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 1.4879,
      "step": 2100
    },
    {
      "epoch": 1.53,
      "eval_loss": 1.5049669742584229,
      "eval_runtime": 556.064,
      "eval_samples_per_second": 11.283,
      "eval_steps_per_second": 1.412,
      "step": 2100
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 1.4477,
      "step": 2200
    },
    {
      "epoch": 1.6,
      "eval_loss": 1.47505521774292,
      "eval_runtime": 555.9558,
      "eval_samples_per_second": 11.285,
      "eval_steps_per_second": 1.412,
      "step": 2200
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002,
      "loss": 1.4289,
      "step": 2300
    },
    {
      "epoch": 1.68,
      "eval_loss": 1.4537893533706665,
      "eval_runtime": 556.1711,
      "eval_samples_per_second": 11.281,
      "eval_steps_per_second": 1.411,
      "step": 2300
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0002,
      "loss": 1.4179,
      "step": 2400
    },
    {
      "epoch": 1.75,
      "eval_loss": 1.4315266609191895,
      "eval_runtime": 556.1848,
      "eval_samples_per_second": 11.28,
      "eval_steps_per_second": 1.411,
      "step": 2400
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 1.3847,
      "step": 2500
    },
    {
      "epoch": 1.82,
      "eval_loss": 1.4086532592773438,
      "eval_runtime": 556.2903,
      "eval_samples_per_second": 11.278,
      "eval_steps_per_second": 1.411,
      "step": 2500
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0002,
      "loss": 1.3664,
      "step": 2600
    },
    {
      "epoch": 1.89,
      "eval_loss": 1.3867732286453247,
      "eval_runtime": 556.1989,
      "eval_samples_per_second": 11.28,
      "eval_steps_per_second": 1.411,
      "step": 2600
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0002,
      "loss": 1.3493,
      "step": 2700
    },
    {
      "epoch": 1.97,
      "eval_loss": 1.361178994178772,
      "eval_runtime": 556.384,
      "eval_samples_per_second": 11.276,
      "eval_steps_per_second": 1.411,
      "step": 2700
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0002,
      "loss": 1.2768,
      "step": 2800
    },
    {
      "epoch": 2.04,
      "eval_loss": 1.3487073183059692,
      "eval_runtime": 556.4859,
      "eval_samples_per_second": 11.274,
      "eval_steps_per_second": 1.411,
      "step": 2800
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0002,
      "loss": 1.2273,
      "step": 2900
    },
    {
      "epoch": 2.11,
      "eval_loss": 1.3275220394134521,
      "eval_runtime": 556.9562,
      "eval_samples_per_second": 11.265,
      "eval_steps_per_second": 1.409,
      "step": 2900
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0002,
      "loss": 1.2451,
      "step": 3000
    },
    {
      "epoch": 2.19,
      "eval_loss": 1.307883858680725,
      "eval_runtime": 555.8716,
      "eval_samples_per_second": 11.287,
      "eval_steps_per_second": 1.412,
      "step": 3000
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0002,
      "loss": 1.208,
      "step": 3100
    },
    {
      "epoch": 2.26,
      "eval_loss": 1.2866222858428955,
      "eval_runtime": 556.1597,
      "eval_samples_per_second": 11.281,
      "eval_steps_per_second": 1.411,
      "step": 3100
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0002,
      "loss": 1.2102,
      "step": 3200
    },
    {
      "epoch": 2.33,
      "eval_loss": 1.27306067943573,
      "eval_runtime": 765.8818,
      "eval_samples_per_second": 8.192,
      "eval_steps_per_second": 1.025,
      "step": 3200
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0002,
      "loss": 1.1778,
      "step": 3300
    },
    {
      "epoch": 2.4,
      "eval_loss": 1.2513889074325562,
      "eval_runtime": 556.1919,
      "eval_samples_per_second": 11.28,
      "eval_steps_per_second": 1.411,
      "step": 3300
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0002,
      "loss": 1.1666,
      "step": 3400
    },
    {
      "epoch": 2.48,
      "eval_loss": 1.2403820753097534,
      "eval_runtime": 555.8669,
      "eval_samples_per_second": 11.287,
      "eval_steps_per_second": 1.412,
      "step": 3400
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0002,
      "loss": 1.1521,
      "step": 3500
    },
    {
      "epoch": 2.55,
      "eval_loss": 1.2223913669586182,
      "eval_runtime": 556.0099,
      "eval_samples_per_second": 11.284,
      "eval_steps_per_second": 1.412,
      "step": 3500
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0002,
      "loss": 1.1431,
      "step": 3600
    },
    {
      "epoch": 2.62,
      "eval_loss": 1.2031002044677734,
      "eval_runtime": 556.2202,
      "eval_samples_per_second": 11.28,
      "eval_steps_per_second": 1.411,
      "step": 3600
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0002,
      "loss": 1.1138,
      "step": 3700
    },
    {
      "epoch": 2.7,
      "eval_loss": 1.1843565702438354,
      "eval_runtime": 556.3313,
      "eval_samples_per_second": 11.277,
      "eval_steps_per_second": 1.411,
      "step": 3700
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0002,
      "loss": 1.1002,
      "step": 3800
    },
    {
      "epoch": 2.77,
      "eval_loss": 1.1716701984405518,
      "eval_runtime": 556.4009,
      "eval_samples_per_second": 11.276,
      "eval_steps_per_second": 1.411,
      "step": 3800
    },
    {
      "epoch": 2.84,
      "learning_rate": 0.0002,
      "loss": 1.0883,
      "step": 3900
    },
    {
      "epoch": 2.84,
      "eval_loss": 1.1533775329589844,
      "eval_runtime": 556.2301,
      "eval_samples_per_second": 11.28,
      "eval_steps_per_second": 1.411,
      "step": 3900
    },
    {
      "epoch": 2.91,
      "learning_rate": 0.0002,
      "loss": 1.0899,
      "step": 4000
    },
    {
      "epoch": 2.91,
      "eval_loss": 1.139776587486267,
      "eval_runtime": 556.1925,
      "eval_samples_per_second": 11.28,
      "eval_steps_per_second": 1.411,
      "step": 4000
    },
    {
      "epoch": 2.99,
      "learning_rate": 0.0002,
      "loss": 1.0699,
      "step": 4100
    },
    {
      "epoch": 2.99,
      "eval_loss": 1.1207506656646729,
      "eval_runtime": 916.3506,
      "eval_samples_per_second": 6.847,
      "eval_steps_per_second": 0.857,
      "step": 4100
    }
  ],
  "logging_steps": 100,
  "max_steps": 4116,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 1.0672136421373379e+18,
  "trial_name": null,
  "trial_params": null
}