trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 6498,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.23084025854108955,
      "grad_norm": 10.19813346862793,
      "learning_rate": 4.615266235764851e-05,
      "loss": 2.8709,
      "step": 500
    },
    {
      "epoch": 0.4616805170821791,
      "grad_norm": 8.437438011169434,
      "learning_rate": 4.230532471529702e-05,
      "loss": 2.0854,
      "step": 1000
    },
    {
      "epoch": 0.6925207756232687,
      "grad_norm": 8.001243591308594,
      "learning_rate": 3.845798707294552e-05,
      "loss": 1.846,
      "step": 1500
    },
    {
      "epoch": 0.9233610341643582,
      "grad_norm": 7.907439231872559,
      "learning_rate": 3.4610649430594035e-05,
      "loss": 1.6927,
      "step": 2000
    },
    {
      "epoch": 1.154201292705448,
      "grad_norm": 6.441333293914795,
      "learning_rate": 3.076331178824254e-05,
      "loss": 1.5576,
      "step": 2500
    },
    {
      "epoch": 1.3850415512465375,
      "grad_norm": 7.415187358856201,
      "learning_rate": 2.6915974145891044e-05,
      "loss": 1.4918,
      "step": 3000
    },
    {
      "epoch": 1.615881809787627,
      "grad_norm": 7.552712440490723,
      "learning_rate": 2.3068636503539552e-05,
      "loss": 1.4271,
      "step": 3500
    },
    {
      "epoch": 1.8467220683287167,
      "grad_norm": 7.328059196472168,
      "learning_rate": 1.922129886118806e-05,
      "loss": 1.3894,
      "step": 4000
    },
    {
      "epoch": 2.0775623268698062,
      "grad_norm": 7.679264068603516,
      "learning_rate": 1.5373961218836565e-05,
      "loss": 1.339,
      "step": 4500
    },
    {
      "epoch": 2.308402585410896,
      "grad_norm": 7.513192653656006,
      "learning_rate": 1.1526623576485073e-05,
      "loss": 1.2782,
      "step": 5000
    },
    {
      "epoch": 2.5392428439519854,
      "grad_norm": 7.99367094039917,
      "learning_rate": 7.67928593413358e-06,
      "loss": 1.2609,
      "step": 5500
    },
    {
      "epoch": 2.770083102493075,
      "grad_norm": 7.764072418212891,
      "learning_rate": 3.831948291782087e-06,
      "loss": 1.2275,
      "step": 6000
    },
    {
      "epoch": 3.0,
      "step": 6498,
      "total_flos": 4.846865511470285e+16,
      "train_loss": 1.5912532506997346,
      "train_runtime": 107034.0122,
      "train_samples_per_second": 0.486,
      "train_steps_per_second": 0.061
    }
  ],
  "logging_steps": 500,
  "max_steps": 6498,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.846865511470285e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}