trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 4120,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.24271844660194175,
      "grad_norm": 0.29302138090133667,
      "learning_rate": 9.393203883495146e-05,
      "loss": 2.8318,
      "step": 250
    },
    {
      "epoch": 0.4854368932038835,
      "grad_norm": 0.24823348224163055,
      "learning_rate": 8.786407766990292e-05,
      "loss": 2.4178,
      "step": 500
    },
    {
      "epoch": 0.7281553398058253,
      "grad_norm": 0.2736216187477112,
      "learning_rate": 8.179611650485438e-05,
      "loss": 2.3486,
      "step": 750
    },
    {
      "epoch": 0.970873786407767,
      "grad_norm": 0.23155324161052704,
      "learning_rate": 7.572815533980583e-05,
      "loss": 2.321,
      "step": 1000
    },
    {
      "epoch": 1.2135922330097086,
      "grad_norm": 0.3087535500526428,
      "learning_rate": 6.966019417475728e-05,
      "loss": 2.3124,
      "step": 1250
    },
    {
      "epoch": 1.4563106796116505,
      "grad_norm": 0.2022610604763031,
      "learning_rate": 6.359223300970875e-05,
      "loss": 2.307,
      "step": 1500
    },
    {
      "epoch": 1.6990291262135924,
      "grad_norm": 0.1987154483795166,
      "learning_rate": 5.752427184466019e-05,
      "loss": 2.3019,
      "step": 1750
    },
    {
      "epoch": 1.941747572815534,
      "grad_norm": 0.16502627730369568,
      "learning_rate": 5.145631067961165e-05,
      "loss": 2.3073,
      "step": 2000
    },
    {
      "epoch": 2.1844660194174756,
      "grad_norm": 0.19471481442451477,
      "learning_rate": 4.538834951456311e-05,
      "loss": 2.2879,
      "step": 2250
    },
    {
      "epoch": 2.4271844660194173,
      "grad_norm": 0.20921674370765686,
      "learning_rate": 3.9320388349514564e-05,
      "loss": 2.2888,
      "step": 2500
    },
    {
      "epoch": 2.6699029126213594,
      "grad_norm": 0.2216738611459732,
      "learning_rate": 3.325242718446602e-05,
      "loss": 2.2997,
      "step": 2750
    },
    {
      "epoch": 2.912621359223301,
      "grad_norm": 0.2476879507303238,
      "learning_rate": 2.7184466019417475e-05,
      "loss": 2.2748,
      "step": 3000
    },
    {
      "epoch": 3.1553398058252426,
      "grad_norm": 0.2942667603492737,
      "learning_rate": 2.111650485436893e-05,
      "loss": 2.2749,
      "step": 3250
    },
    {
      "epoch": 3.3980582524271843,
      "grad_norm": 0.2994736135005951,
      "learning_rate": 1.5048543689320387e-05,
      "loss": 2.2738,
      "step": 3500
    },
    {
      "epoch": 3.6407766990291264,
      "grad_norm": 0.2753419876098633,
      "learning_rate": 8.980582524271845e-06,
      "loss": 2.2557,
      "step": 3750
    },
    {
      "epoch": 3.883495145631068,
      "grad_norm": 0.277048796415329,
      "learning_rate": 2.912621359223301e-06,
      "loss": 2.2512,
      "step": 4000
    },
    {
      "epoch": 4.0,
      "step": 4120,
      "total_flos": 7.617568236584731e+18,
      "train_loss": 2.3324776880949445,
      "train_runtime": 61881.1241,
      "train_samples_per_second": 8.522,
      "train_steps_per_second": 0.067
    }
  ],
  "logging_steps": 250,
  "max_steps": 4120,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 7.617568236584731e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}