{
  "best_metric": 0.03224072605371475,
  "best_model_checkpoint": "/gs/fs/tgh-25IAK/ue02946/PLANT/Season_based_split_performance/final_full_250304_ver2_12_fixed_rp__s20000_lr1.00E-04_wd0.01_rwd0.005_csew0_csea0_sw0.2_csewv0_swv0.2_cw0.05_rid256_dp0.05_eid64_edp0.1_lgw0.001_bs16_facebookesm2_t33_650M_UR50D/trained_until_full/results/checkpoint-19000",
  "epoch": 9.336609336609337,
  "eval_steps": 1000,
  "global_step": 19000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2457002457002457,
      "grad_norm": 0.9868232607841492,
      "learning_rate": 1.2425e-05,
      "loss": 0.1679,
      "step": 500
    },
    {
      "epoch": 0.4914004914004914,
      "grad_norm": 0.6529730558395386,
      "learning_rate": 2.4925000000000003e-05,
      "loss": 0.0803,
      "step": 1000
    },
    {
      "epoch": 0.4914004914004914,
      "eval_loss": 0.06003545597195625,
      "eval_runtime": 174.4345,
      "eval_samples_per_second": 40.869,
      "eval_steps_per_second": 1.278,
      "step": 1000
    },
    {
      "epoch": 0.7371007371007371,
      "grad_norm": 0.24933239817619324,
      "learning_rate": 3.7425e-05,
      "loss": 0.0655,
      "step": 1500
    },
    {
      "epoch": 0.9828009828009828,
      "grad_norm": 0.7596633434295654,
      "learning_rate": 4.992500000000001e-05,
      "loss": 0.0544,
      "step": 2000
    },
    {
      "epoch": 0.9828009828009828,
      "eval_loss": 0.05782468616962433,
      "eval_runtime": 174.3269,
      "eval_samples_per_second": 40.894,
      "eval_steps_per_second": 1.279,
      "step": 2000
    },
    {
      "epoch": 1.2285012285012284,
      "grad_norm": 0.31257227063179016,
      "learning_rate": 4.8619444444444446e-05,
      "loss": 0.0454,
      "step": 2500
    },
    {
      "epoch": 1.4742014742014742,
      "grad_norm": 0.21421055495738983,
      "learning_rate": 4.723055555555555e-05,
      "loss": 0.043,
      "step": 3000
    },
    {
      "epoch": 1.4742014742014742,
      "eval_loss": 0.043084025382995605,
      "eval_runtime": 174.7024,
      "eval_samples_per_second": 40.807,
      "eval_steps_per_second": 1.276,
      "step": 3000
    },
    {
      "epoch": 1.71990171990172,
      "grad_norm": 0.26165878772735596,
      "learning_rate": 4.5841666666666665e-05,
      "loss": 0.0417,
      "step": 3500
    },
    {
      "epoch": 1.9656019656019657,
      "grad_norm": 0.3407607674598694,
      "learning_rate": 4.4452777777777785e-05,
      "loss": 0.0394,
      "step": 4000
    },
    {
      "epoch": 1.9656019656019657,
      "eval_loss": 0.037903204560279846,
      "eval_runtime": 175.2535,
      "eval_samples_per_second": 40.678,
      "eval_steps_per_second": 1.272,
      "step": 4000
    },
    {
      "epoch": 2.211302211302211,
      "grad_norm": 0.41331663727760315,
      "learning_rate": 4.306388888888889e-05,
      "loss": 0.0355,
      "step": 4500
    },
    {
      "epoch": 2.457002457002457,
      "grad_norm": 0.3084980845451355,
      "learning_rate": 4.1675e-05,
      "loss": 0.0372,
      "step": 5000
    },
    {
      "epoch": 2.457002457002457,
      "eval_loss": 0.042647846043109894,
      "eval_runtime": 175.1611,
      "eval_samples_per_second": 40.7,
      "eval_steps_per_second": 1.273,
      "step": 5000
    },
    {
      "epoch": 2.7027027027027026,
      "grad_norm": 0.1795072853565216,
      "learning_rate": 4.0286111111111116e-05,
      "loss": 0.0346,
      "step": 5500
    },
    {
      "epoch": 2.9484029484029484,
      "grad_norm": 0.2850644588470459,
      "learning_rate": 3.889722222222222e-05,
      "loss": 0.0329,
      "step": 6000
    },
    {
      "epoch": 2.9484029484029484,
      "eval_loss": 0.03940274938941002,
      "eval_runtime": 175.0589,
      "eval_samples_per_second": 40.723,
      "eval_steps_per_second": 1.274,
      "step": 6000
    },
    {
      "epoch": 3.194103194103194,
      "grad_norm": 0.196182519197464,
      "learning_rate": 3.7508333333333335e-05,
      "loss": 0.033,
      "step": 6500
    },
    {
      "epoch": 3.43980343980344,
      "grad_norm": 0.17266607284545898,
      "learning_rate": 3.611944444444445e-05,
      "loss": 0.0312,
      "step": 7000
    },
    {
      "epoch": 3.43980343980344,
      "eval_loss": 0.037528131157159805,
      "eval_runtime": 175.0252,
      "eval_samples_per_second": 40.731,
      "eval_steps_per_second": 1.274,
      "step": 7000
    },
    {
      "epoch": 3.6855036855036856,
      "grad_norm": 0.18219518661499023,
      "learning_rate": 3.4730555555555553e-05,
      "loss": 0.0312,
      "step": 7500
    },
    {
      "epoch": 3.9312039312039313,
      "grad_norm": 0.32914480566978455,
      "learning_rate": 3.3341666666666666e-05,
      "loss": 0.0311,
      "step": 8000
    },
    {
      "epoch": 3.9312039312039313,
      "eval_loss": 0.036644209176301956,
      "eval_runtime": 174.8439,
      "eval_samples_per_second": 40.774,
      "eval_steps_per_second": 1.275,
      "step": 8000
    },
    {
      "epoch": 4.176904176904177,
      "grad_norm": 0.1908116638660431,
      "learning_rate": 3.195277777777778e-05,
      "loss": 0.0294,
      "step": 8500
    },
    {
      "epoch": 4.422604422604422,
      "grad_norm": 0.23989616334438324,
      "learning_rate": 3.056388888888889e-05,
      "loss": 0.029,
      "step": 9000
    },
    {
      "epoch": 4.422604422604422,
      "eval_loss": 0.03631925955414772,
      "eval_runtime": 174.7772,
      "eval_samples_per_second": 40.789,
      "eval_steps_per_second": 1.276,
      "step": 9000
    },
    {
      "epoch": 4.6683046683046685,
      "grad_norm": 0.15896788239479065,
      "learning_rate": 2.9175e-05,
      "loss": 0.0294,
      "step": 9500
    },
    {
      "epoch": 4.914004914004914,
      "grad_norm": 0.17607934772968292,
      "learning_rate": 2.7786111111111114e-05,
      "loss": 0.0278,
      "step": 10000
    },
    {
      "epoch": 4.914004914004914,
      "eval_loss": 0.0335356779396534,
      "eval_runtime": 174.7415,
      "eval_samples_per_second": 40.797,
      "eval_steps_per_second": 1.276,
      "step": 10000
    },
    {
      "epoch": 5.15970515970516,
      "grad_norm": 0.1342485249042511,
      "learning_rate": 2.64e-05,
      "loss": 0.0285,
      "step": 10500
    },
    {
      "epoch": 5.405405405405405,
      "grad_norm": 0.32292836904525757,
      "learning_rate": 2.5011111111111114e-05,
      "loss": 0.0275,
      "step": 11000
    },
    {
      "epoch": 5.405405405405405,
      "eval_loss": 0.03467724099755287,
      "eval_runtime": 175.0323,
      "eval_samples_per_second": 40.73,
      "eval_steps_per_second": 1.274,
      "step": 11000
    },
    {
      "epoch": 5.651105651105651,
      "grad_norm": 0.4820247292518616,
      "learning_rate": 2.3622222222222223e-05,
      "loss": 0.028,
      "step": 11500
    },
    {
      "epoch": 5.896805896805897,
      "grad_norm": 0.29012107849121094,
      "learning_rate": 2.2233333333333333e-05,
      "loss": 0.0258,
      "step": 12000
    },
    {
      "epoch": 5.896805896805897,
      "eval_loss": 0.033297277987003326,
      "eval_runtime": 175.0904,
      "eval_samples_per_second": 40.716,
      "eval_steps_per_second": 1.274,
      "step": 12000
    },
    {
      "epoch": 6.142506142506143,
      "grad_norm": 0.1986035406589508,
      "learning_rate": 2.0847222222222224e-05,
      "loss": 0.026,
      "step": 12500
    },
    {
      "epoch": 6.388206388206388,
      "grad_norm": 0.258876770734787,
      "learning_rate": 1.9458333333333333e-05,
      "loss": 0.0254,
      "step": 13000
    },
    {
      "epoch": 6.388206388206388,
      "eval_loss": 0.034665919840335846,
      "eval_runtime": 174.8409,
      "eval_samples_per_second": 40.774,
      "eval_steps_per_second": 1.275,
      "step": 13000
    },
    {
      "epoch": 6.6339066339066335,
      "grad_norm": 0.6466019153594971,
      "learning_rate": 1.8069444444444446e-05,
      "loss": 0.026,
      "step": 13500
    },
    {
      "epoch": 6.87960687960688,
      "grad_norm": 0.23270685970783234,
      "learning_rate": 1.668055555555556e-05,
      "loss": 0.0256,
      "step": 14000
    },
    {
      "epoch": 6.87960687960688,
      "eval_loss": 0.03323034569621086,
      "eval_runtime": 174.9701,
      "eval_samples_per_second": 40.744,
      "eval_steps_per_second": 1.275,
      "step": 14000
    },
    {
      "epoch": 7.125307125307125,
      "grad_norm": 0.6487894058227539,
      "learning_rate": 1.5291666666666668e-05,
      "loss": 0.0254,
      "step": 14500
    },
    {
      "epoch": 7.371007371007371,
      "grad_norm": 0.5228617191314697,
      "learning_rate": 1.3905555555555555e-05,
      "loss": 0.0247,
      "step": 15000
    },
    {
      "epoch": 7.371007371007371,
      "eval_loss": 0.03286266699433327,
      "eval_runtime": 174.9609,
      "eval_samples_per_second": 40.746,
      "eval_steps_per_second": 1.275,
      "step": 15000
    },
    {
      "epoch": 7.616707616707616,
      "grad_norm": 0.4490284323692322,
      "learning_rate": 1.2516666666666668e-05,
      "loss": 0.0253,
      "step": 15500
    },
    {
      "epoch": 7.862407862407863,
      "grad_norm": 0.40585392713546753,
      "learning_rate": 1.112777777777778e-05,
      "loss": 0.0239,
      "step": 16000
    },
    {
      "epoch": 7.862407862407863,
      "eval_loss": 0.03392437845468521,
      "eval_runtime": 174.9232,
      "eval_samples_per_second": 40.755,
      "eval_steps_per_second": 1.275,
      "step": 16000
    },
    {
      "epoch": 8.108108108108109,
      "grad_norm": 0.14088092744350433,
      "learning_rate": 9.738888888888889e-06,
      "loss": 0.0238,
      "step": 16500
    },
    {
      "epoch": 8.353808353808354,
      "grad_norm": 0.10795734822750092,
      "learning_rate": 8.350000000000001e-06,
      "loss": 0.0241,
      "step": 17000
    },
    {
      "epoch": 8.353808353808354,
      "eval_loss": 0.033751752227544785,
      "eval_runtime": 174.9375,
      "eval_samples_per_second": 40.752,
      "eval_steps_per_second": 1.275,
      "step": 17000
    },
    {
      "epoch": 8.5995085995086,
      "grad_norm": 0.20282748341560364,
      "learning_rate": 6.963888888888889e-06,
      "loss": 0.0223,
      "step": 17500
    },
    {
      "epoch": 8.845208845208845,
      "grad_norm": 0.1708468645811081,
      "learning_rate": 5.575e-06,
      "loss": 0.0233,
      "step": 18000
    },
    {
      "epoch": 8.845208845208845,
      "eval_loss": 0.03310641273856163,
      "eval_runtime": 175.167,
      "eval_samples_per_second": 40.698,
      "eval_steps_per_second": 1.273,
      "step": 18000
    },
    {
      "epoch": 9.090909090909092,
      "grad_norm": 0.18372096121311188,
      "learning_rate": 4.186111111111112e-06,
      "loss": 0.0232,
      "step": 18500
    },
    {
      "epoch": 9.336609336609337,
      "grad_norm": 0.4115198254585266,
      "learning_rate": 2.797222222222222e-06,
      "loss": 0.0228,
      "step": 19000
    },
    {
      "epoch": 9.336609336609337,
      "eval_loss": 0.03224072605371475,
      "eval_runtime": 175.1142,
      "eval_samples_per_second": 40.711,
      "eval_steps_per_second": 1.273,
      "step": 19000
    }
  ],
  "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
}