last-checkpoint/trainer_state.json

{
  "best_metric": 0.6268890955591723,
  "best_model_checkpoint": "modernbert-medical-classifier/checkpoint-644",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 736,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_f1": 0.2555762427123016,
      "eval_loss": 1.1453216075897217,
      "eval_runtime": 7.2906,
      "eval_samples_per_second": 12.619,
      "eval_steps_per_second": 3.155,
      "step": 92
    },
    {
      "epoch": 1.0869565217391304,
      "grad_norm": 27.595176696777344,
      "learning_rate": 8.695652173913044e-06,
      "loss": 1.2432,
      "step": 100
    },
    {
      "epoch": 2.0,
      "eval_f1": 0.22693092661056924,
      "eval_loss": 1.1348090171813965,
      "eval_runtime": 7.2682,
      "eval_samples_per_second": 12.658,
      "eval_steps_per_second": 3.164,
      "step": 184
    },
    {
      "epoch": 2.1739130434782608,
      "grad_norm": 10.296772003173828,
      "learning_rate": 1.739130434782609e-05,
      "loss": 1.1447,
      "step": 200
    },
    {
      "epoch": 3.0,
      "eval_f1": 0.2623751468860165,
      "eval_loss": 1.194385051727295,
      "eval_runtime": 7.2397,
      "eval_samples_per_second": 12.708,
      "eval_steps_per_second": 3.177,
      "step": 276
    },
    {
      "epoch": 3.260869565217391,
      "grad_norm": 8.205018997192383,
      "learning_rate": 1.932367149758454e-05,
      "loss": 1.0924,
      "step": 300
    },
    {
      "epoch": 4.0,
      "eval_f1": 0.46725195094760313,
      "eval_loss": 0.9543380737304688,
      "eval_runtime": 7.257,
      "eval_samples_per_second": 12.677,
      "eval_steps_per_second": 3.169,
      "step": 368
    },
    {
      "epoch": 4.3478260869565215,
      "grad_norm": 11.217053413391113,
      "learning_rate": 1.8357487922705315e-05,
      "loss": 0.9918,
      "step": 400
    },
    {
      "epoch": 5.0,
      "eval_f1": 0.48917610522469857,
      "eval_loss": 1.0969688892364502,
      "eval_runtime": 7.2942,
      "eval_samples_per_second": 12.613,
      "eval_steps_per_second": 3.153,
      "step": 460
    },
    {
      "epoch": 5.434782608695652,
      "grad_norm": 17.887250900268555,
      "learning_rate": 1.739130434782609e-05,
      "loss": 0.8981,
      "step": 500
    },
    {
      "epoch": 6.0,
      "eval_f1": 0.5668988939449298,
      "eval_loss": 1.2105615139007568,
      "eval_runtime": 7.3182,
      "eval_samples_per_second": 12.571,
      "eval_steps_per_second": 3.143,
      "step": 552
    },
    {
      "epoch": 6.521739130434782,
      "grad_norm": 26.001848220825195,
      "learning_rate": 1.6425120772946863e-05,
      "loss": 0.9261,
      "step": 600
    },
    {
      "epoch": 7.0,
      "eval_f1": 0.6268890955591723,
      "eval_loss": 0.8952301740646362,
      "eval_runtime": 7.2679,
      "eval_samples_per_second": 12.658,
      "eval_steps_per_second": 3.165,
      "step": 644
    },
    {
      "epoch": 7.608695652173913,
      "grad_norm": 21.24846649169922,
      "learning_rate": 1.5458937198067633e-05,
      "loss": 0.8208,
      "step": 700
    },
    {
      "epoch": 8.0,
      "eval_f1": 0.5480383816552855,
      "eval_loss": 1.1408698558807373,
      "eval_runtime": 7.2908,
      "eval_samples_per_second": 12.619,
      "eval_steps_per_second": 3.155,
      "step": 736
    }
  ],
  "logging_steps": 100,
  "max_steps": 2300,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 25,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5052834964944000.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}