trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9667673716012084,
  "eval_steps": 500,
  "global_step": 186,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16112789526686808,
      "grad_norm": 0.8870864510536194,
      "learning_rate": 9.473684210526315e-06,
      "loss": 0.9173,
      "step": 10
    },
    {
      "epoch": 0.32225579053373615,
      "grad_norm": 0.9371799230575562,
      "learning_rate": 2e-05,
      "loss": 0.7161,
      "step": 20
    },
    {
      "epoch": 0.48338368580060426,
      "grad_norm": 1.0169328451156616,
      "learning_rate": 1.982357681294114e-05,
      "loss": 0.5601,
      "step": 30
    },
    {
      "epoch": 0.6445115810674723,
      "grad_norm": 1.0059415102005005,
      "learning_rate": 1.930053227995095e-05,
      "loss": 0.3905,
      "step": 40
    },
    {
      "epoch": 0.8056394763343404,
      "grad_norm": 0.7323524355888367,
      "learning_rate": 1.8449321837726207e-05,
      "loss": 0.2684,
      "step": 50
    },
    {
      "epoch": 0.9667673716012085,
      "grad_norm": 0.6315535306930542,
      "learning_rate": 1.729998013808192e-05,
      "loss": 0.1756,
      "step": 60
    },
    {
      "epoch": 1.1127895266868078,
      "grad_norm": 0.6027315855026245,
      "learning_rate": 1.5893061286152275e-05,
      "loss": 0.129,
      "step": 70
    },
    {
      "epoch": 1.2739174219536757,
      "grad_norm": 0.6192404627799988,
      "learning_rate": 1.4278207903495388e-05,
      "loss": 0.0491,
      "step": 80
    },
    {
      "epoch": 1.4350453172205437,
      "grad_norm": 0.5543455481529236,
      "learning_rate": 1.251239950619149e-05,
      "loss": 0.0381,
      "step": 90
    },
    {
      "epoch": 1.596173212487412,
      "grad_norm": 0.5133563280105591,
      "learning_rate": 1.0657942003278107e-05,
      "loss": 0.0392,
      "step": 100
    },
    {
      "epoch": 1.75730110775428,
      "grad_norm": 0.29823583364486694,
      "learning_rate": 8.78026925534108e-06,
      "loss": 0.0262,
      "step": 110
    },
    {
      "epoch": 1.918429003021148,
      "grad_norm": 0.38609743118286133,
      "learning_rate": 6.945634264469338e-06,
      "loss": 0.0218,
      "step": 120
    },
    {
      "epoch": 2.0644511581067473,
      "grad_norm": 0.1530793011188507,
      "learning_rate": 5.218771461098733e-06,
      "loss": 0.01,
      "step": 130
    },
    {
      "epoch": 2.2255790533736155,
      "grad_norm": 0.25707682967185974,
      "learning_rate": 3.6606125731060815e-06,
      "loss": 0.0052,
      "step": 140
    },
    {
      "epoch": 2.3867069486404833,
      "grad_norm": 0.08537214994430542,
      "learning_rate": 2.3261366718841307e-06,
      "loss": 0.0067,
      "step": 150
    },
    {
      "epoch": 2.5478348439073515,
      "grad_norm": 0.1156155988574028,
      "learning_rate": 1.2624302557422475e-06,
      "loss": 0.0043,
      "step": 160
    },
    {
      "epoch": 2.7089627391742197,
      "grad_norm": 0.14377349615097046,
      "learning_rate": 5.070258198865741e-07,
      "loss": 0.0041,
      "step": 170
    },
    {
      "epoch": 2.8700906344410875,
      "grad_norm": 0.12491826713085175,
      "learning_rate": 8.657753593552142e-08,
      "loss": 0.005,
      "step": 180
    },
    {
      "epoch": 2.9667673716012084,
      "step": 186,
      "total_flos": 3.483502879052923e+17,
      "train_loss": 0.1811602087090573,
      "train_runtime": 2439.3943,
      "train_samples_per_second": 4.882,
      "train_steps_per_second": 0.076
    }
  ],
  "logging_steps": 10,
  "max_steps": 186,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.483502879052923e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}