{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.827586206896552,
  "eval_steps": 2,
  "global_step": 35,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14,
      "learning_rate": 0.0001,
      "loss": 0.7533,
      "step": 1
    },
    {
      "epoch": 0.28,
      "learning_rate": 9.978670881475172e-05,
      "loss": 0.8025,
      "step": 2
    },
    {
      "epoch": 0.28,
      "eval_loss": 0.6696242094039917,
      "eval_runtime": 44.2051,
      "eval_samples_per_second": 1.968,
      "eval_steps_per_second": 1.968,
      "step": 2
    },
    {
      "epoch": 0.41,
      "learning_rate": 9.91486549841951e-05,
      "loss": 0.6626,
      "step": 3
    },
    {
      "epoch": 0.55,
      "learning_rate": 9.809128215864097e-05,
      "loss": 0.6912,
      "step": 4
    },
    {
      "epoch": 0.55,
      "eval_loss": 0.6283016800880432,
      "eval_runtime": 44.1458,
      "eval_samples_per_second": 1.971,
      "eval_steps_per_second": 1.971,
      "step": 4
    },
    {
      "epoch": 0.69,
      "learning_rate": 9.662361147021779e-05,
      "loss": 0.5708,
      "step": 5
    },
    {
      "epoch": 0.83,
      "learning_rate": 9.475816456775313e-05,
      "loss": 0.6832,
      "step": 6
    },
    {
      "epoch": 0.83,
      "eval_loss": 0.6122540235519409,
      "eval_runtime": 44.1996,
      "eval_samples_per_second": 1.968,
      "eval_steps_per_second": 1.968,
      "step": 6
    },
    {
      "epoch": 0.97,
      "learning_rate": 9.251085678648072e-05,
      "loss": 0.601,
      "step": 7
    },
    {
      "epoch": 1.1,
      "learning_rate": 8.9900861364012e-05,
      "loss": 0.6154,
      "step": 8
    },
    {
      "epoch": 1.1,
      "eval_loss": 0.6039846539497375,
      "eval_runtime": 44.163,
      "eval_samples_per_second": 1.97,
      "eval_steps_per_second": 1.97,
      "step": 8
    },
    {
      "epoch": 1.24,
      "learning_rate": 8.695044586103296e-05,
      "loss": 0.5479,
      "step": 9
    },
    {
      "epoch": 1.38,
      "learning_rate": 8.368478218232787e-05,
      "loss": 0.5438,
      "step": 10
    },
    {
      "epoch": 1.38,
      "eval_loss": 0.5960671901702881,
      "eval_runtime": 44.0267,
      "eval_samples_per_second": 1.976,
      "eval_steps_per_second": 1.976,
      "step": 10
    },
    {
      "epoch": 1.52,
      "learning_rate": 8.013173181896283e-05,
      "loss": 0.6171,
      "step": 11
    },
    {
      "epoch": 1.66,
      "learning_rate": 7.63216081438678e-05,
      "loss": 0.5781,
      "step": 12
    },
    {
      "epoch": 1.66,
      "eval_loss": 0.5879073143005371,
      "eval_runtime": 43.9085,
      "eval_samples_per_second": 1.981,
      "eval_steps_per_second": 1.981,
      "step": 12
    },
    {
      "epoch": 1.79,
      "learning_rate": 7.228691778882693e-05,
      "loss": 0.5811,
      "step": 13
    },
    {
      "epoch": 1.93,
      "learning_rate": 6.806208330935766e-05,
      "loss": 0.4899,
      "step": 14
    },
    {
      "epoch": 1.93,
      "eval_loss": 0.5821757912635803,
      "eval_runtime": 43.9108,
      "eval_samples_per_second": 1.981,
      "eval_steps_per_second": 1.981,
      "step": 14
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.368314950360415e-05,
      "loss": 0.4649,
      "step": 15
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.918747589082853e-05,
      "loss": 0.5615,
      "step": 16
    },
    {
      "epoch": 2.21,
      "eval_loss": 0.5788118839263916,
      "eval_runtime": 43.9641,
      "eval_samples_per_second": 1.979,
      "eval_steps_per_second": 1.979,
      "step": 16
    },
    {
      "epoch": 2.34,
      "learning_rate": 5.4613417973165106e-05,
      "loss": 0.4919,
      "step": 17
    },
    {
      "epoch": 2.48,
      "learning_rate": 5e-05,
      "loss": 0.4755,
      "step": 18
    },
    {
      "epoch": 2.48,
      "eval_loss": 0.5790432691574097,
      "eval_runtime": 43.9953,
      "eval_samples_per_second": 1.977,
      "eval_steps_per_second": 1.977,
      "step": 18
    },
    {
      "epoch": 2.62,
      "learning_rate": 4.5386582026834906e-05,
      "loss": 0.4525,
      "step": 19
    },
    {
      "epoch": 2.76,
      "learning_rate": 4.0812524109171476e-05,
      "loss": 0.4665,
      "step": 20
    },
    {
      "epoch": 2.76,
      "eval_loss": 0.5807910561561584,
      "eval_runtime": 44.0437,
      "eval_samples_per_second": 1.975,
      "eval_steps_per_second": 1.975,
      "step": 20
    },
    {
      "epoch": 2.9,
      "learning_rate": 3.631685049639586e-05,
      "loss": 0.5079,
      "step": 21
    },
    {
      "epoch": 3.03,
      "learning_rate": 3.1937916690642356e-05,
      "loss": 0.4525,
      "step": 22
    },
    {
      "epoch": 3.03,
      "eval_loss": 0.5798379182815552,
      "eval_runtime": 44.0127,
      "eval_samples_per_second": 1.977,
      "eval_steps_per_second": 1.977,
      "step": 22
    },
    {
      "epoch": 3.17,
      "learning_rate": 2.771308221117309e-05,
      "loss": 0.4697,
      "step": 23
    },
    {
      "epoch": 3.31,
      "learning_rate": 2.3678391856132204e-05,
      "loss": 0.4508,
      "step": 24
    },
    {
      "epoch": 3.31,
      "eval_loss": 0.5790860056877136,
      "eval_runtime": 44.0414,
      "eval_samples_per_second": 1.975,
      "eval_steps_per_second": 1.975,
      "step": 24
    },
    {
      "epoch": 3.45,
      "learning_rate": 1.9868268181037185e-05,
      "loss": 0.4566,
      "step": 25
    },
    {
      "epoch": 3.59,
      "learning_rate": 1.631521781767214e-05,
      "loss": 0.4934,
      "step": 26
    },
    {
      "epoch": 3.59,
      "eval_loss": 0.5790104269981384,
      "eval_runtime": 44.0612,
      "eval_samples_per_second": 1.975,
      "eval_steps_per_second": 1.975,
      "step": 26
    },
    {
      "epoch": 3.72,
      "learning_rate": 1.3049554138967051e-05,
      "loss": 0.3934,
      "step": 27
    },
    {
      "epoch": 3.86,
      "learning_rate": 1.0099138635988026e-05,
      "loss": 0.4291,
      "step": 28
    },
    {
      "epoch": 3.86,
      "eval_loss": 0.5794057846069336,
      "eval_runtime": 44.0534,
      "eval_samples_per_second": 1.975,
      "eval_steps_per_second": 1.975,
      "step": 28
    },
    {
      "epoch": 4.0,
      "learning_rate": 7.489143213519301e-06,
      "loss": 0.4485,
      "step": 29
    },
    {
      "epoch": 4.14,
      "learning_rate": 5.241835432246889e-06,
      "loss": 0.4624,
      "step": 30
    },
    {
      "epoch": 4.14,
      "eval_loss": 0.5792465806007385,
      "eval_runtime": 44.0238,
      "eval_samples_per_second": 1.976,
      "eval_steps_per_second": 1.976,
      "step": 30
    },
    {
      "epoch": 4.28,
      "learning_rate": 3.376388529782215e-06,
      "loss": 0.449,
      "step": 31
    },
    {
      "epoch": 4.41,
      "learning_rate": 1.908717841359048e-06,
      "loss": 0.4353,
      "step": 32
    },
    {
      "epoch": 4.41,
      "eval_loss": 0.579724907875061,
      "eval_runtime": 43.9242,
      "eval_samples_per_second": 1.981,
      "eval_steps_per_second": 1.981,
      "step": 32
    },
    {
      "epoch": 4.55,
      "learning_rate": 8.513450158049108e-07,
      "loss": 0.4108,
      "step": 33
    },
    {
      "epoch": 4.69,
      "learning_rate": 2.1329118524827662e-07,
      "loss": 0.4546,
      "step": 34
    },
    {
      "epoch": 4.69,
      "eval_loss": 0.5798984169960022,
      "eval_runtime": 43.9248,
      "eval_samples_per_second": 1.981,
      "eval_steps_per_second": 1.981,
      "step": 34
    },
    {
      "epoch": 4.83,
      "learning_rate": 0.0,
      "loss": 0.4189,
      "step": 35
    },
    {
      "epoch": 4.83,
      "step": 35,
      "total_flos": 9.3962984585429e+16,
      "train_loss": 0.5252525525433677,
      "train_runtime": 2623.1378,
      "train_samples_per_second": 0.66,
      "train_steps_per_second": 0.013
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 35,
  "num_train_epochs": 5,
  "save_steps": 4,
  "total_flos": 9.3962984585429e+16,
  "trial_name": null,
  "trial_params": null
}