{
  "best_metric": 0.10928191244602203,
  "best_model_checkpoint": "/media/user/Expansion/flan-t5-small-untokenizer2/checkpoint-22500",
  "epoch": 1.0,
  "eval_steps": 2500,
  "global_step": 25000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "grad_norm": 0.623769998550415,
      "learning_rate": 4.9e-05,
      "loss": 0.2187,
      "num_input_tokens_seen": 1550704,
      "step": 500
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.6147541403770447,
      "learning_rate": 4.8e-05,
      "loss": 0.1616,
      "num_input_tokens_seen": 3113636,
      "step": 1000
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.8041962385177612,
      "learning_rate": 4.7e-05,
      "loss": 0.1465,
      "num_input_tokens_seen": 4634732,
      "step": 1500
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.6499223709106445,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.1523,
      "num_input_tokens_seen": 6206076,
      "step": 2000
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.6867215633392334,
      "learning_rate": 4.5e-05,
      "loss": 0.1477,
      "num_input_tokens_seen": 7758904,
      "step": 2500
    },
    {
      "epoch": 0.1,
      "eval_loss": 0.11968539655208588,
      "eval_runtime": 45.394,
      "eval_samples_per_second": 55.073,
      "eval_steps_per_second": 6.895,
      "num_input_tokens_seen": 7758904,
      "step": 2500
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.5227354764938354,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.1422,
      "num_input_tokens_seen": 9286000,
      "step": 3000
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.6782410144805908,
      "learning_rate": 4.3e-05,
      "loss": 0.1417,
      "num_input_tokens_seen": 10857236,
      "step": 3500
    },
    {
      "epoch": 0.16,
      "grad_norm": 2.188730001449585,
      "learning_rate": 4.2e-05,
      "loss": 0.1296,
      "num_input_tokens_seen": 12388284,
      "step": 4000
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.5791600346565247,
      "learning_rate": 4.1e-05,
      "loss": 0.1345,
      "num_input_tokens_seen": 13970100,
      "step": 4500
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.1668968200683594,
      "learning_rate": 4e-05,
      "loss": 0.1331,
      "num_input_tokens_seen": 15535000,
      "step": 5000
    },
    {
      "epoch": 0.2,
      "eval_loss": 0.11603260785341263,
      "eval_runtime": 45.3659,
      "eval_samples_per_second": 55.107,
      "eval_steps_per_second": 6.899,
      "num_input_tokens_seen": 15535000,
      "step": 5000
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.5479309558868408,
      "learning_rate": 3.9000000000000006e-05,
      "loss": 0.1319,
      "num_input_tokens_seen": 17098792,
      "step": 5500
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.6089518666267395,
      "learning_rate": 3.8e-05,
      "loss": 0.1325,
      "num_input_tokens_seen": 18665500,
      "step": 6000
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.9072483777999878,
      "learning_rate": 3.7e-05,
      "loss": 0.1327,
      "num_input_tokens_seen": 20242492,
      "step": 6500
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.6144304275512695,
      "learning_rate": 3.6e-05,
      "loss": 0.1317,
      "num_input_tokens_seen": 21840216,
      "step": 7000
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.6446326375007629,
      "learning_rate": 3.5e-05,
      "loss": 0.1295,
      "num_input_tokens_seen": 23379996,
      "step": 7500
    },
    {
      "epoch": 0.3,
      "eval_loss": 0.11423994600772858,
      "eval_runtime": 45.4091,
      "eval_samples_per_second": 55.055,
      "eval_steps_per_second": 6.893,
      "num_input_tokens_seen": 23379996,
      "step": 7500
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.35889774560928345,
      "learning_rate": 3.4000000000000007e-05,
      "loss": 0.1279,
      "num_input_tokens_seen": 24947356,
      "step": 8000
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.6276640892028809,
      "learning_rate": 3.3e-05,
      "loss": 0.1296,
      "num_input_tokens_seen": 26524048,
      "step": 8500
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.16105327010154724,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 0.1257,
      "num_input_tokens_seen": 28056324,
      "step": 9000
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.7733196020126343,
      "learning_rate": 3.1e-05,
      "loss": 0.1198,
      "num_input_tokens_seen": 29608912,
      "step": 9500
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.6086151599884033,
      "learning_rate": 3e-05,
      "loss": 0.1257,
      "num_input_tokens_seen": 31164076,
      "step": 10000
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.1128210574388504,
      "eval_runtime": 45.589,
      "eval_samples_per_second": 54.838,
      "eval_steps_per_second": 6.866,
      "num_input_tokens_seen": 31164076,
      "step": 10000
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.40136536955833435,
      "learning_rate": 2.9e-05,
      "loss": 0.131,
      "num_input_tokens_seen": 32749448,
      "step": 10500
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.9361308217048645,
      "learning_rate": 2.8000000000000003e-05,
      "loss": 0.114,
      "num_input_tokens_seen": 34266344,
      "step": 11000
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.6654324531555176,
      "learning_rate": 2.7000000000000002e-05,
      "loss": 0.1401,
      "num_input_tokens_seen": 35845312,
      "step": 11500
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.22417037189006805,
      "learning_rate": 2.6000000000000002e-05,
      "loss": 0.1183,
      "num_input_tokens_seen": 37392204,
      "step": 12000
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.354526549577713,
      "learning_rate": 2.5e-05,
      "loss": 0.1148,
      "num_input_tokens_seen": 38943032,
      "step": 12500
    },
    {
      "epoch": 0.5,
      "eval_loss": 0.11148922145366669,
      "eval_runtime": 45.5906,
      "eval_samples_per_second": 54.836,
      "eval_steps_per_second": 6.865,
      "num_input_tokens_seen": 38943032,
      "step": 12500
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5592128038406372,
      "learning_rate": 2.4e-05,
      "loss": 0.1237,
      "num_input_tokens_seen": 40522284,
      "step": 13000
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5023311376571655,
      "learning_rate": 2.3000000000000003e-05,
      "loss": 0.1188,
      "num_input_tokens_seen": 42078660,
      "step": 13500
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.528414249420166,
      "learning_rate": 2.2000000000000003e-05,
      "loss": 0.1306,
      "num_input_tokens_seen": 43633484,
      "step": 14000
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.7868823409080505,
      "learning_rate": 2.1e-05,
      "loss": 0.1167,
      "num_input_tokens_seen": 45163904,
      "step": 14500
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.30285605788230896,
      "learning_rate": 2e-05,
      "loss": 0.1219,
      "num_input_tokens_seen": 46747616,
      "step": 15000
    },
    {
      "epoch": 0.6,
      "eval_loss": 0.11066804826259613,
      "eval_runtime": 45.6215,
      "eval_samples_per_second": 54.799,
      "eval_steps_per_second": 6.861,
      "num_input_tokens_seen": 46747616,
      "step": 15000
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.18638424575328827,
      "learning_rate": 1.9e-05,
      "loss": 0.1295,
      "num_input_tokens_seen": 48319848,
      "step": 15500
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.22312051057815552,
      "learning_rate": 1.8e-05,
      "loss": 0.116,
      "num_input_tokens_seen": 49864448,
      "step": 16000
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.3801974356174469,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 0.1321,
      "num_input_tokens_seen": 51461076,
      "step": 16500
    },
    {
      "epoch": 0.68,
      "grad_norm": 2.719569206237793,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.1172,
      "num_input_tokens_seen": 53024800,
      "step": 17000
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.3563300669193268,
      "learning_rate": 1.5e-05,
      "loss": 0.1112,
      "num_input_tokens_seen": 54513880,
      "step": 17500
    },
    {
      "epoch": 0.7,
      "eval_loss": 0.11027801036834717,
      "eval_runtime": 45.5331,
      "eval_samples_per_second": 54.905,
      "eval_steps_per_second": 6.874,
      "num_input_tokens_seen": 54513880,
      "step": 17500
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.36904799938201904,
      "learning_rate": 1.4000000000000001e-05,
      "loss": 0.1228,
      "num_input_tokens_seen": 56077248,
      "step": 18000
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.20697401463985443,
      "learning_rate": 1.3000000000000001e-05,
      "loss": 0.1228,
      "num_input_tokens_seen": 57639976,
      "step": 18500
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.25590789318084717,
      "learning_rate": 1.2e-05,
      "loss": 0.1193,
      "num_input_tokens_seen": 59184096,
      "step": 19000
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.8957023620605469,
      "learning_rate": 1.1000000000000001e-05,
      "loss": 0.1233,
      "num_input_tokens_seen": 60758920,
      "step": 19500
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.3526408076286316,
      "learning_rate": 1e-05,
      "loss": 0.1161,
      "num_input_tokens_seen": 62302092,
      "step": 20000
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.10997864603996277,
      "eval_runtime": 45.5162,
      "eval_samples_per_second": 54.925,
      "eval_steps_per_second": 6.877,
      "num_input_tokens_seen": 62302092,
      "step": 20000
    },
    {
      "epoch": 0.82,
      "grad_norm": 1.0477633476257324,
      "learning_rate": 9e-06,
      "loss": 0.1143,
      "num_input_tokens_seen": 63829116,
      "step": 20500
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.30011287331581116,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.1187,
      "num_input_tokens_seen": 65361664,
      "step": 21000
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.3339782953262329,
      "learning_rate": 7.000000000000001e-06,
      "loss": 0.1149,
      "num_input_tokens_seen": 66920576,
      "step": 21500
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.2934107780456543,
      "learning_rate": 6e-06,
      "loss": 0.1153,
      "num_input_tokens_seen": 68429876,
      "step": 22000
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.3181125521659851,
      "learning_rate": 5e-06,
      "loss": 0.1215,
      "num_input_tokens_seen": 70000044,
      "step": 22500
    },
    {
      "epoch": 0.9,
      "eval_loss": 0.10928191244602203,
      "eval_runtime": 45.5863,
      "eval_samples_per_second": 54.841,
      "eval_steps_per_second": 6.866,
      "num_input_tokens_seen": 70000044,
      "step": 22500
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5395247936248779,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.119,
      "num_input_tokens_seen": 71571012,
      "step": 23000
    },
    {
      "epoch": 0.94,
      "grad_norm": 1.1183191537857056,
      "learning_rate": 3e-06,
      "loss": 0.1196,
      "num_input_tokens_seen": 73134988,
      "step": 23500
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.40107282996177673,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.1107,
      "num_input_tokens_seen": 74666536,
      "step": 24000
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.1762804388999939,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.1096,
      "num_input_tokens_seen": 76182928,
      "step": 24500
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.2036857306957245,
      "learning_rate": 0.0,
      "loss": 0.1207,
      "num_input_tokens_seen": 77750728,
      "step": 25000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.10940483212471008,
      "eval_runtime": 45.4609,
      "eval_samples_per_second": 54.992,
      "eval_steps_per_second": 6.885,
      "num_input_tokens_seen": 77750728,
      "step": 25000
    },
    {
      "epoch": 1.0,
      "num_input_tokens_seen": 77750728,
      "step": 25000,
      "total_flos": 2.8228754041061376e+16,
      "train_loss": 0.12799154373168944,
      "train_runtime": 4779.3877,
      "train_samples_per_second": 20.923,
      "train_steps_per_second": 5.231,
      "train_tokens_per_second": 16267.926
    }
  ],
  "logging_steps": 500,
  "max_steps": 25000,
  "num_input_tokens_seen": 77750728,
  "num_train_epochs": 1,
  "save_steps": 2500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.8228754041061376e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}