{
  "best_metric": 0.027055351063609123,
  "best_model_checkpoint": "mgh6/TCS_Pairing/checkpoint-25000",
  "epoch": 2.9836496001909536,
  "eval_steps": 1000,
  "global_step": 25000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12,
      "learning_rate": 9.602180053307873e-05,
      "loss": 0.0929,
      "step": 1000
    },
    {
      "epoch": 0.12,
      "eval_loss": 0.0718587264418602,
      "eval_runtime": 142.087,
      "eval_samples_per_second": 178.778,
      "eval_steps_per_second": 22.353,
      "step": 1000
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.204360106615746e-05,
      "loss": 0.0654,
      "step": 2000
    },
    {
      "epoch": 0.24,
      "eval_loss": 0.06243297830224037,
      "eval_runtime": 142.08,
      "eval_samples_per_second": 178.787,
      "eval_steps_per_second": 22.354,
      "step": 2000
    },
    {
      "epoch": 0.36,
      "learning_rate": 8.806540159923618e-05,
      "loss": 0.056,
      "step": 3000
    },
    {
      "epoch": 0.36,
      "eval_loss": 0.05325521528720856,
      "eval_runtime": 142.1931,
      "eval_samples_per_second": 178.644,
      "eval_steps_per_second": 22.336,
      "step": 3000
    },
    {
      "epoch": 0.48,
      "learning_rate": 8.408720213231491e-05,
      "loss": 0.0494,
      "step": 4000
    },
    {
      "epoch": 0.48,
      "eval_loss": 0.0536990687251091,
      "eval_runtime": 142.149,
      "eval_samples_per_second": 178.7,
      "eval_steps_per_second": 22.343,
      "step": 4000
    },
    {
      "epoch": 0.6,
      "learning_rate": 8.010900266539365e-05,
      "loss": 0.0456,
      "step": 5000
    },
    {
      "epoch": 0.6,
      "eval_loss": 0.04832467809319496,
      "eval_runtime": 141.7755,
      "eval_samples_per_second": 179.171,
      "eval_steps_per_second": 22.402,
      "step": 5000
    },
    {
      "epoch": 0.72,
      "learning_rate": 7.613080319847238e-05,
      "loss": 0.0423,
      "step": 6000
    },
    {
      "epoch": 0.72,
      "eval_loss": 0.043777432292699814,
      "eval_runtime": 142.1073,
      "eval_samples_per_second": 178.752,
      "eval_steps_per_second": 22.349,
      "step": 6000
    },
    {
      "epoch": 0.84,
      "learning_rate": 7.21526037315511e-05,
      "loss": 0.0392,
      "step": 7000
    },
    {
      "epoch": 0.84,
      "eval_loss": 0.04013779014348984,
      "eval_runtime": 142.123,
      "eval_samples_per_second": 178.733,
      "eval_steps_per_second": 22.347,
      "step": 7000
    },
    {
      "epoch": 0.95,
      "learning_rate": 6.817440426462983e-05,
      "loss": 0.0383,
      "step": 8000
    },
    {
      "epoch": 0.95,
      "eval_loss": 0.04025523364543915,
      "eval_runtime": 142.0835,
      "eval_samples_per_second": 178.782,
      "eval_steps_per_second": 22.353,
      "step": 8000
    },
    {
      "epoch": 1.07,
      "learning_rate": 6.419620479770856e-05,
      "loss": 0.0308,
      "step": 9000
    },
    {
      "epoch": 1.07,
      "eval_loss": 0.03931814804673195,
      "eval_runtime": 142.1394,
      "eval_samples_per_second": 178.712,
      "eval_steps_per_second": 22.344,
      "step": 9000
    },
    {
      "epoch": 1.19,
      "learning_rate": 6.021800533078728e-05,
      "loss": 0.0279,
      "step": 10000
    },
    {
      "epoch": 1.19,
      "eval_loss": 0.037444014102220535,
      "eval_runtime": 142.0806,
      "eval_samples_per_second": 178.786,
      "eval_steps_per_second": 22.354,
      "step": 10000
    },
    {
      "epoch": 1.31,
      "learning_rate": 5.6239805863866016e-05,
      "loss": 0.0272,
      "step": 11000
    },
    {
      "epoch": 1.31,
      "eval_loss": 0.037211526185274124,
      "eval_runtime": 142.1393,
      "eval_samples_per_second": 178.712,
      "eval_steps_per_second": 22.344,
      "step": 11000
    },
    {
      "epoch": 1.43,
      "learning_rate": 5.226160639694474e-05,
      "loss": 0.0262,
      "step": 12000
    },
    {
      "epoch": 1.43,
      "eval_loss": 0.03615099936723709,
      "eval_runtime": 142.2278,
      "eval_samples_per_second": 178.601,
      "eval_steps_per_second": 22.33,
      "step": 12000
    },
    {
      "epoch": 1.55,
      "learning_rate": 4.8283406930023476e-05,
      "loss": 0.0245,
      "step": 13000
    },
    {
      "epoch": 1.55,
      "eval_loss": 0.03587919846177101,
      "eval_runtime": 142.4557,
      "eval_samples_per_second": 178.315,
      "eval_steps_per_second": 22.295,
      "step": 13000
    },
    {
      "epoch": 1.67,
      "learning_rate": 4.43052074631022e-05,
      "loss": 0.0243,
      "step": 14000
    },
    {
      "epoch": 1.67,
      "eval_loss": 0.0344834141433239,
      "eval_runtime": 142.1512,
      "eval_samples_per_second": 178.697,
      "eval_steps_per_second": 22.342,
      "step": 14000
    },
    {
      "epoch": 1.79,
      "learning_rate": 4.032700799618093e-05,
      "loss": 0.0226,
      "step": 15000
    },
    {
      "epoch": 1.79,
      "eval_loss": 0.03409423679113388,
      "eval_runtime": 142.2177,
      "eval_samples_per_second": 178.613,
      "eval_steps_per_second": 22.332,
      "step": 15000
    },
    {
      "epoch": 1.91,
      "learning_rate": 3.634880852925966e-05,
      "loss": 0.0231,
      "step": 16000
    },
    {
      "epoch": 1.91,
      "eval_loss": 0.03197489306330681,
      "eval_runtime": 140.7435,
      "eval_samples_per_second": 180.484,
      "eval_steps_per_second": 22.566,
      "step": 16000
    },
    {
      "epoch": 2.03,
      "learning_rate": 3.237060906233839e-05,
      "loss": 0.0198,
      "step": 17000
    },
    {
      "epoch": 2.03,
      "eval_loss": 0.03260968625545502,
      "eval_runtime": 142.0512,
      "eval_samples_per_second": 178.823,
      "eval_steps_per_second": 22.358,
      "step": 17000
    },
    {
      "epoch": 2.15,
      "learning_rate": 2.839240959541711e-05,
      "loss": 0.013,
      "step": 18000
    },
    {
      "epoch": 2.15,
      "eval_loss": 0.03267335891723633,
      "eval_runtime": 141.695,
      "eval_samples_per_second": 179.272,
      "eval_steps_per_second": 22.414,
      "step": 18000
    },
    {
      "epoch": 2.27,
      "learning_rate": 2.441421012849584e-05,
      "loss": 0.0132,
      "step": 19000
    },
    {
      "epoch": 2.27,
      "eval_loss": 0.03128109499812126,
      "eval_runtime": 142.0978,
      "eval_samples_per_second": 178.764,
      "eval_steps_per_second": 22.351,
      "step": 19000
    },
    {
      "epoch": 2.39,
      "learning_rate": 2.043601066157457e-05,
      "loss": 0.0123,
      "step": 20000
    },
    {
      "epoch": 2.39,
      "eval_loss": 0.03134540468454361,
      "eval_runtime": 142.1226,
      "eval_samples_per_second": 178.733,
      "eval_steps_per_second": 22.347,
      "step": 20000
    },
    {
      "epoch": 2.51,
      "learning_rate": 1.64578111946533e-05,
      "loss": 0.0123,
      "step": 21000
    },
    {
      "epoch": 2.51,
      "eval_loss": 0.030218619853258133,
      "eval_runtime": 141.8386,
      "eval_samples_per_second": 179.091,
      "eval_steps_per_second": 22.392,
      "step": 21000
    },
    {
      "epoch": 2.63,
      "learning_rate": 1.247961172773203e-05,
      "loss": 0.012,
      "step": 22000
    },
    {
      "epoch": 2.63,
      "eval_loss": 0.028801346197724342,
      "eval_runtime": 142.0191,
      "eval_samples_per_second": 178.863,
      "eval_steps_per_second": 22.363,
      "step": 22000
    },
    {
      "epoch": 2.74,
      "learning_rate": 8.501412260810758e-06,
      "loss": 0.0116,
      "step": 23000
    },
    {
      "epoch": 2.74,
      "eval_loss": 0.027987737208604813,
      "eval_runtime": 141.3339,
      "eval_samples_per_second": 179.73,
      "eval_steps_per_second": 22.472,
      "step": 23000
    },
    {
      "epoch": 2.86,
      "learning_rate": 4.523212793889486e-06,
      "loss": 0.0107,
      "step": 24000
    },
    {
      "epoch": 2.86,
      "eval_loss": 0.027928264811635017,
      "eval_runtime": 141.818,
      "eval_samples_per_second": 179.117,
      "eval_steps_per_second": 22.395,
      "step": 24000
    },
    {
      "epoch": 2.98,
      "learning_rate": 5.450133269682142e-07,
      "loss": 0.0115,
      "step": 25000
    },
    {
      "epoch": 2.98,
      "eval_loss": 0.027055351063609123,
      "eval_runtime": 142.8238,
      "eval_samples_per_second": 177.856,
      "eval_steps_per_second": 22.237,
      "step": 25000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 25137,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1000,
  "total_flos": 6.252295780604789e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}