{
  "best_metric": 1.6118786334991455,
  "best_model_checkpoint": "../../experiments_checkpoints/MAdAiLab/FacebookAI/roberta_base_scotus/checkpoint-200",
  "epoch": 3.0,
  "eval_steps": 50,
  "global_step": 237,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.13,
      "grad_norm": 4.399165153503418,
      "learning_rate": 1.9156118143459917e-05,
      "loss": 2.4218,
      "step": 10
    },
    {
      "epoch": 0.25,
      "grad_norm": 3.196627378463745,
      "learning_rate": 1.8312236286919833e-05,
      "loss": 2.1889,
      "step": 20
    },
    {
      "epoch": 0.38,
      "grad_norm": 2.2070045471191406,
      "learning_rate": 1.746835443037975e-05,
      "loss": 2.1449,
      "step": 30
    },
    {
      "epoch": 0.51,
      "grad_norm": 3.898432493209839,
      "learning_rate": 1.662447257383966e-05,
      "loss": 2.0514,
      "step": 40
    },
    {
      "epoch": 0.63,
      "grad_norm": 5.449430465698242,
      "learning_rate": 1.578059071729958e-05,
      "loss": 1.9275,
      "step": 50
    },
    {
      "epoch": 0.63,
      "eval_accuracy": 0.41,
      "eval_f1_macro": 0.11253090508032011,
      "eval_f1_micro": 0.41,
      "eval_loss": 1.915439486503601,
      "eval_runtime": 1.2656,
      "eval_samples_per_second": 1106.191,
      "eval_steps_per_second": 17.383,
      "step": 50
    },
    {
      "epoch": 0.76,
      "grad_norm": 4.739038467407227,
      "learning_rate": 1.4936708860759495e-05,
      "loss": 1.8774,
      "step": 60
    },
    {
      "epoch": 0.89,
      "grad_norm": 5.2211127281188965,
      "learning_rate": 1.4092827004219412e-05,
      "loss": 1.7642,
      "step": 70
    },
    {
      "epoch": 1.01,
      "grad_norm": 5.152080535888672,
      "learning_rate": 1.3248945147679326e-05,
      "loss": 1.8278,
      "step": 80
    },
    {
      "epoch": 1.14,
      "grad_norm": 6.538599967956543,
      "learning_rate": 1.240506329113924e-05,
      "loss": 1.6386,
      "step": 90
    },
    {
      "epoch": 1.27,
      "grad_norm": 8.518353462219238,
      "learning_rate": 1.1561181434599158e-05,
      "loss": 1.5742,
      "step": 100
    },
    {
      "epoch": 1.27,
      "eval_accuracy": 0.4707142857142857,
      "eval_f1_macro": 0.17818280427876487,
      "eval_f1_micro": 0.4707142857142857,
      "eval_loss": 1.7446955442428589,
      "eval_runtime": 1.3224,
      "eval_samples_per_second": 1058.65,
      "eval_steps_per_second": 16.636,
      "step": 100
    },
    {
      "epoch": 1.39,
      "grad_norm": 6.387670516967773,
      "learning_rate": 1.0717299578059072e-05,
      "loss": 1.6975,
      "step": 110
    },
    {
      "epoch": 1.52,
      "grad_norm": 7.155932903289795,
      "learning_rate": 9.87341772151899e-06,
      "loss": 1.6592,
      "step": 120
    },
    {
      "epoch": 1.65,
      "grad_norm": 8.153914451599121,
      "learning_rate": 9.029535864978903e-06,
      "loss": 1.6166,
      "step": 130
    },
    {
      "epoch": 1.77,
      "grad_norm": 7.831178188323975,
      "learning_rate": 8.18565400843882e-06,
      "loss": 1.5245,
      "step": 140
    },
    {
      "epoch": 1.9,
      "grad_norm": 7.43508768081665,
      "learning_rate": 7.341772151898735e-06,
      "loss": 1.6563,
      "step": 150
    },
    {
      "epoch": 1.9,
      "eval_accuracy": 0.4785714285714286,
      "eval_f1_macro": 0.20760404191283174,
      "eval_f1_micro": 0.4785714285714286,
      "eval_loss": 1.6859331130981445,
      "eval_runtime": 1.2827,
      "eval_samples_per_second": 1091.43,
      "eval_steps_per_second": 17.151,
      "step": 150
    },
    {
      "epoch": 2.03,
      "grad_norm": 8.414676666259766,
      "learning_rate": 6.49789029535865e-06,
      "loss": 1.5582,
      "step": 160
    },
    {
      "epoch": 2.15,
      "grad_norm": 8.495677947998047,
      "learning_rate": 5.654008438818566e-06,
      "loss": 1.4579,
      "step": 170
    },
    {
      "epoch": 2.28,
      "grad_norm": 7.206579685211182,
      "learning_rate": 4.8101265822784815e-06,
      "loss": 1.5167,
      "step": 180
    },
    {
      "epoch": 2.41,
      "grad_norm": 6.6391167640686035,
      "learning_rate": 3.9662447257383965e-06,
      "loss": 1.4687,
      "step": 190
    },
    {
      "epoch": 2.53,
      "grad_norm": 5.48373556137085,
      "learning_rate": 3.1223628691983127e-06,
      "loss": 1.4743,
      "step": 200
    },
    {
      "epoch": 2.53,
      "eval_accuracy": 0.5135714285714286,
      "eval_f1_macro": 0.24550969494780733,
      "eval_f1_micro": 0.5135714285714286,
      "eval_loss": 1.6118786334991455,
      "eval_runtime": 1.2892,
      "eval_samples_per_second": 1085.913,
      "eval_steps_per_second": 17.064,
      "step": 200
    },
    {
      "epoch": 2.66,
      "grad_norm": 7.982447147369385,
      "learning_rate": 2.278481012658228e-06,
      "loss": 1.481,
      "step": 210
    },
    {
      "epoch": 2.78,
      "grad_norm": 7.101485729217529,
      "learning_rate": 1.4345991561181436e-06,
      "loss": 1.4405,
      "step": 220
    },
    {
      "epoch": 2.91,
      "grad_norm": 9.05599308013916,
      "learning_rate": 5.907172995780591e-07,
      "loss": 1.4664,
      "step": 230
    },
    {
      "epoch": 3.0,
      "step": 237,
      "total_flos": 997815706714112.0,
      "train_loss": 1.706272036717411,
      "train_runtime": 56.9364,
      "train_samples_per_second": 263.452,
      "train_steps_per_second": 4.163
    }
  ],
  "logging_steps": 10,
  "max_steps": 237,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 50,
  "total_flos": 997815706714112.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}