{
  "best_metric": 0.034771159291267395,
  "best_model_checkpoint": "/gs/fs/tgh-25IAK/ue02946/PLANT/Season_based_split_performance/final_full_250304_ver2_12__s20000_lr1.00E-04_wd0.01_rwd0.005_csew0_csea0_sw0.2_csewv0_swv0.2_cw0.05_rid256_dp0.05_eid64_edp0.1_lgw0.01_bs16_facebookesm2_t33_650M_UR50D/trained_until_full/results/checkpoint-16000",
  "epoch": 8.0,
  "eval_steps": 1000,
  "global_step": 16000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.25,
      "grad_norm": 1.430809736251831,
      "learning_rate": 1.2475e-05,
      "loss": 0.1385,
      "step": 500
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.6614887118339539,
      "learning_rate": 2.4975e-05,
      "loss": 0.0693,
      "step": 1000
    },
    {
      "epoch": 0.5,
      "eval_loss": 0.05656217411160469,
      "eval_runtime": 175.6555,
      "eval_samples_per_second": 40.585,
      "eval_steps_per_second": 1.27,
      "step": 1000
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.6037147641181946,
      "learning_rate": 3.7475e-05,
      "loss": 0.0612,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.4972088038921356,
      "learning_rate": 4.9975e-05,
      "loss": 0.0577,
      "step": 2000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.06209351494908333,
      "eval_runtime": 175.8577,
      "eval_samples_per_second": 40.538,
      "eval_steps_per_second": 1.268,
      "step": 2000
    },
    {
      "epoch": 1.25,
      "grad_norm": 1.0353070497512817,
      "learning_rate": 4.861388888888889e-05,
      "loss": 0.0511,
      "step": 2500
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.8515058755874634,
      "learning_rate": 4.7225e-05,
      "loss": 0.0447,
      "step": 3000
    },
    {
      "epoch": 1.5,
      "eval_loss": 0.050771959125995636,
      "eval_runtime": 175.8405,
      "eval_samples_per_second": 40.542,
      "eval_steps_per_second": 1.268,
      "step": 3000
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.5539710521697998,
      "learning_rate": 4.5836111111111115e-05,
      "loss": 0.0441,
      "step": 3500
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.7269946336746216,
      "learning_rate": 4.444722222222222e-05,
      "loss": 0.0409,
      "step": 4000
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.0459725446999073,
      "eval_runtime": 175.8069,
      "eval_samples_per_second": 40.55,
      "eval_steps_per_second": 1.268,
      "step": 4000
    },
    {
      "epoch": 2.25,
      "grad_norm": 0.25292640924453735,
      "learning_rate": 4.3058333333333334e-05,
      "loss": 0.0385,
      "step": 4500
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.2268402874469757,
      "learning_rate": 4.166944444444445e-05,
      "loss": 0.04,
      "step": 5000
    },
    {
      "epoch": 2.5,
      "eval_loss": 0.04381516948342323,
      "eval_runtime": 175.8092,
      "eval_samples_per_second": 40.55,
      "eval_steps_per_second": 1.268,
      "step": 5000
    },
    {
      "epoch": 2.75,
      "grad_norm": 0.2693457305431366,
      "learning_rate": 4.0283333333333334e-05,
      "loss": 0.0372,
      "step": 5500
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.383963406085968,
      "learning_rate": 3.889444444444445e-05,
      "loss": 0.0359,
      "step": 6000
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.04170890897512436,
      "eval_runtime": 175.8179,
      "eval_samples_per_second": 40.548,
      "eval_steps_per_second": 1.268,
      "step": 6000
    },
    {
      "epoch": 3.25,
      "grad_norm": 0.341470867395401,
      "learning_rate": 3.750555555555555e-05,
      "loss": 0.0366,
      "step": 6500
    },
    {
      "epoch": 3.5,
      "grad_norm": 0.3173449635505676,
      "learning_rate": 3.611666666666667e-05,
      "loss": 0.0342,
      "step": 7000
    },
    {
      "epoch": 3.5,
      "eval_loss": 0.04057968407869339,
      "eval_runtime": 175.921,
      "eval_samples_per_second": 40.524,
      "eval_steps_per_second": 1.268,
      "step": 7000
    },
    {
      "epoch": 3.75,
      "grad_norm": 0.17745588719844818,
      "learning_rate": 3.472777777777778e-05,
      "loss": 0.0334,
      "step": 7500
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.737916886806488,
      "learning_rate": 3.333888888888889e-05,
      "loss": 0.0338,
      "step": 8000
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.03727322071790695,
      "eval_runtime": 175.854,
      "eval_samples_per_second": 40.539,
      "eval_steps_per_second": 1.268,
      "step": 8000
    },
    {
      "epoch": 4.25,
      "grad_norm": 0.4044702351093292,
      "learning_rate": 3.1950000000000004e-05,
      "loss": 0.0315,
      "step": 8500
    },
    {
      "epoch": 4.5,
      "grad_norm": 0.2551765441894531,
      "learning_rate": 3.056388888888889e-05,
      "loss": 0.0325,
      "step": 9000
    },
    {
      "epoch": 4.5,
      "eval_loss": 0.03956814482808113,
      "eval_runtime": 175.8578,
      "eval_samples_per_second": 40.538,
      "eval_steps_per_second": 1.268,
      "step": 9000
    },
    {
      "epoch": 4.75,
      "grad_norm": 0.4832671284675598,
      "learning_rate": 2.9175e-05,
      "loss": 0.0336,
      "step": 9500
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.219984769821167,
      "learning_rate": 2.7786111111111114e-05,
      "loss": 0.0311,
      "step": 10000
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.037063486874103546,
      "eval_runtime": 175.9259,
      "eval_samples_per_second": 40.523,
      "eval_steps_per_second": 1.268,
      "step": 10000
    },
    {
      "epoch": 5.25,
      "grad_norm": 0.1559722125530243,
      "learning_rate": 2.6397222222222223e-05,
      "loss": 0.0303,
      "step": 10500
    },
    {
      "epoch": 5.5,
      "grad_norm": 0.1364031732082367,
      "learning_rate": 2.5008333333333332e-05,
      "loss": 0.0299,
      "step": 11000
    },
    {
      "epoch": 5.5,
      "eval_loss": 0.037668175995349884,
      "eval_runtime": 176.0991,
      "eval_samples_per_second": 40.483,
      "eval_steps_per_second": 1.266,
      "step": 11000
    },
    {
      "epoch": 5.75,
      "grad_norm": 0.20991720259189606,
      "learning_rate": 2.3619444444444445e-05,
      "loss": 0.0308,
      "step": 11500
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.29334667325019836,
      "learning_rate": 2.2230555555555558e-05,
      "loss": 0.0297,
      "step": 12000
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.03977835178375244,
      "eval_runtime": 175.799,
      "eval_samples_per_second": 40.552,
      "eval_steps_per_second": 1.268,
      "step": 12000
    },
    {
      "epoch": 6.25,
      "grad_norm": 0.2716386616230011,
      "learning_rate": 2.0841666666666667e-05,
      "loss": 0.0297,
      "step": 12500
    },
    {
      "epoch": 6.5,
      "grad_norm": 0.30008479952812195,
      "learning_rate": 1.9455555555555558e-05,
      "loss": 0.0284,
      "step": 13000
    },
    {
      "epoch": 6.5,
      "eval_loss": 0.036556895822286606,
      "eval_runtime": 175.7757,
      "eval_samples_per_second": 40.557,
      "eval_steps_per_second": 1.269,
      "step": 13000
    },
    {
      "epoch": 6.75,
      "grad_norm": 0.30487436056137085,
      "learning_rate": 1.8066666666666668e-05,
      "loss": 0.0279,
      "step": 13500
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.2959464490413666,
      "learning_rate": 1.6677777777777777e-05,
      "loss": 0.0288,
      "step": 14000
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.03538262099027634,
      "eval_runtime": 175.7677,
      "eval_samples_per_second": 40.559,
      "eval_steps_per_second": 1.269,
      "step": 14000
    },
    {
      "epoch": 7.25,
      "grad_norm": 0.16555708646774292,
      "learning_rate": 1.528888888888889e-05,
      "loss": 0.0275,
      "step": 14500
    },
    {
      "epoch": 7.5,
      "grad_norm": 0.1510380208492279,
      "learning_rate": 1.3900000000000002e-05,
      "loss": 0.0276,
      "step": 15000
    },
    {
      "epoch": 7.5,
      "eval_loss": 0.03512444719672203,
      "eval_runtime": 175.7946,
      "eval_samples_per_second": 40.553,
      "eval_steps_per_second": 1.269,
      "step": 15000
    },
    {
      "epoch": 7.75,
      "grad_norm": 0.7063078880310059,
      "learning_rate": 1.2513888888888888e-05,
      "loss": 0.0281,
      "step": 15500
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.14875133335590363,
      "learning_rate": 1.1125000000000001e-05,
      "loss": 0.0269,
      "step": 16000
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.034771159291267395,
      "eval_runtime": 175.7842,
      "eval_samples_per_second": 40.555,
      "eval_steps_per_second": 1.269,
      "step": 16000
    }
  ],
  "logging_steps": 500,
  "max_steps": 20000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}