{
  "best_metric": 1.6826781034469604,
  "best_model_checkpoint": "outputs/checkpoint-374",
  "epoch": 10.724014336917563,
  "eval_steps": 500,
  "global_step": 374,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5734767025089605,
      "grad_norm": 1.254808783531189,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 2.9248,
      "step": 20
    },
    {
      "epoch": 0.974910394265233,
      "eval_loss": 2.816770076751709,
      "eval_runtime": 12.0371,
      "eval_samples_per_second": 30.904,
      "eval_steps_per_second": 3.905,
      "step": 34
    },
    {
      "epoch": 1.146953405017921,
      "grad_norm": 1.3472272157669067,
      "learning_rate": 4.000000000000001e-06,
      "loss": 2.8927,
      "step": 40
    },
    {
      "epoch": 1.7204301075268817,
      "grad_norm": 1.56429922580719,
      "learning_rate": 6e-06,
      "loss": 2.827,
      "step": 60
    },
    {
      "epoch": 1.978494623655914,
      "eval_loss": 2.5949196815490723,
      "eval_runtime": 12.0427,
      "eval_samples_per_second": 30.89,
      "eval_steps_per_second": 3.903,
      "step": 69
    },
    {
      "epoch": 2.293906810035842,
      "grad_norm": 2.013805389404297,
      "learning_rate": 8.000000000000001e-06,
      "loss": 2.6619,
      "step": 80
    },
    {
      "epoch": 2.867383512544803,
      "grad_norm": 1.7669492959976196,
      "learning_rate": 1e-05,
      "loss": 2.3052,
      "step": 100
    },
    {
      "epoch": 2.982078853046595,
      "eval_loss": 2.005356550216675,
      "eval_runtime": 12.033,
      "eval_samples_per_second": 30.915,
      "eval_steps_per_second": 3.906,
      "step": 104
    },
    {
      "epoch": 3.4408602150537635,
      "grad_norm": 0.7907297015190125,
      "learning_rate": 9.869113551144754e-06,
      "loss": 2.0129,
      "step": 120
    },
    {
      "epoch": 3.985663082437276,
      "eval_loss": 1.8571571111679077,
      "eval_runtime": 12.0327,
      "eval_samples_per_second": 30.916,
      "eval_steps_per_second": 3.906,
      "step": 139
    },
    {
      "epoch": 4.014336917562724,
      "grad_norm": 0.6714785695075989,
      "learning_rate": 9.48330670957659e-06,
      "loss": 1.9293,
      "step": 140
    },
    {
      "epoch": 4.587813620071684,
      "grad_norm": 0.633492648601532,
      "learning_rate": 8.862778230270276e-06,
      "loss": 1.8755,
      "step": 160
    },
    {
      "epoch": 4.989247311827957,
      "eval_loss": 1.7887212038040161,
      "eval_runtime": 12.0277,
      "eval_samples_per_second": 30.929,
      "eval_steps_per_second": 3.908,
      "step": 174
    },
    {
      "epoch": 5.161290322580645,
      "grad_norm": 0.6426405310630798,
      "learning_rate": 8.04001562085379e-06,
      "loss": 1.8144,
      "step": 180
    },
    {
      "epoch": 5.734767025089606,
      "grad_norm": 0.6501743793487549,
      "learning_rate": 7.058094271806091e-06,
      "loss": 1.7905,
      "step": 200
    },
    {
      "epoch": 5.992831541218638,
      "eval_loss": 1.744368553161621,
      "eval_runtime": 12.0302,
      "eval_samples_per_second": 30.922,
      "eval_steps_per_second": 3.907,
      "step": 209
    },
    {
      "epoch": 6.308243727598566,
      "grad_norm": 0.6884588599205017,
      "learning_rate": 5.968422262499983e-06,
      "loss": 1.7398,
      "step": 220
    },
    {
      "epoch": 6.881720430107527,
      "grad_norm": 0.6701219081878662,
      "learning_rate": 4.82804891282148e-06,
      "loss": 1.7365,
      "step": 240
    },
    {
      "epoch": 6.996415770609319,
      "eval_loss": 1.7139564752578735,
      "eval_runtime": 12.0265,
      "eval_samples_per_second": 30.932,
      "eval_steps_per_second": 3.908,
      "step": 244
    },
    {
      "epoch": 7.455197132616488,
      "grad_norm": 0.6416382789611816,
      "learning_rate": 3.6966779900140193e-06,
      "loss": 1.7276,
      "step": 260
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.6951279640197754,
      "eval_runtime": 12.0247,
      "eval_samples_per_second": 30.936,
      "eval_steps_per_second": 3.909,
      "step": 279
    },
    {
      "epoch": 8.028673835125447,
      "grad_norm": 0.5849926471710205,
      "learning_rate": 2.633541943047334e-06,
      "loss": 1.685,
      "step": 280
    },
    {
      "epoch": 8.602150537634408,
      "grad_norm": 0.7124162912368774,
      "learning_rate": 1.6943008126564164e-06,
      "loss": 1.695,
      "step": 300
    },
    {
      "epoch": 8.974910394265233,
      "eval_loss": 1.6860439777374268,
      "eval_runtime": 12.028,
      "eval_samples_per_second": 30.928,
      "eval_steps_per_second": 3.908,
      "step": 313
    },
    {
      "epoch": 9.175627240143369,
      "grad_norm": 0.6544684767723083,
      "learning_rate": 9.281281733115288e-07,
      "loss": 1.6987,
      "step": 320
    },
    {
      "epoch": 9.74910394265233,
      "grad_norm": 0.8023189902305603,
      "learning_rate": 3.751366714022342e-07,
      "loss": 1.6868,
      "step": 340
    },
    {
      "epoch": 9.978494623655914,
      "eval_loss": 1.6829451322555542,
      "eval_runtime": 12.0275,
      "eval_samples_per_second": 30.929,
      "eval_steps_per_second": 3.908,
      "step": 348
    },
    {
      "epoch": 10.32258064516129,
      "grad_norm": 0.6753117442131042,
      "learning_rate": 6.427794450134529e-08,
      "loss": 1.692,
      "step": 360
    },
    {
      "epoch": 10.724014336917563,
      "eval_loss": 1.6826781034469604,
      "eval_runtime": 12.0235,
      "eval_samples_per_second": 30.939,
      "eval_steps_per_second": 3.909,
      "step": 374
    }
  ],
  "logging_steps": 20,
  "max_steps": 374,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 11,
  "save_steps": 500,
  "total_flos": 1.968170106536755e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}