trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 69,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.014492753623188406,
      "grad_norm": 1.90625,
      "learning_rate": 2.857142857142857e-05,
      "loss": 1.742,
      "step": 1
    },
    {
      "epoch": 0.07246376811594203,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00014285714285714287,
      "loss": 1.6788,
      "step": 5
    },
    {
      "epoch": 0.14492753623188406,
      "grad_norm": 0.69140625,
      "learning_rate": 0.00019884683243281116,
      "loss": 1.2504,
      "step": 10
    },
    {
      "epoch": 0.21739130434782608,
      "grad_norm": 0.216796875,
      "learning_rate": 0.00019189578116202307,
      "loss": 1.0864,
      "step": 15
    },
    {
      "epoch": 0.2898550724637681,
      "grad_norm": 0.22265625,
      "learning_rate": 0.00017907757369376985,
      "loss": 0.9902,
      "step": 20
    },
    {
      "epoch": 0.36231884057971014,
      "grad_norm": 0.2158203125,
      "learning_rate": 0.0001612105982547663,
      "loss": 0.901,
      "step": 25
    },
    {
      "epoch": 0.43478260869565216,
      "grad_norm": 0.181640625,
      "learning_rate": 0.00013943558551133186,
      "loss": 0.854,
      "step": 30
    },
    {
      "epoch": 0.5072463768115942,
      "grad_norm": 0.177734375,
      "learning_rate": 0.00011514277775045768,
      "loss": 0.8177,
      "step": 35
    },
    {
      "epoch": 0.5797101449275363,
      "grad_norm": 0.166015625,
      "learning_rate": 8.98831678012568e-05,
      "loss": 0.7782,
      "step": 40
    },
    {
      "epoch": 0.6521739130434783,
      "grad_norm": 0.169921875,
      "learning_rate": 6.526947471551798e-05,
      "loss": 0.792,
      "step": 45
    },
    {
      "epoch": 0.7246376811594203,
      "grad_norm": 0.1669921875,
      "learning_rate": 4.287317849052075e-05,
      "loss": 0.7844,
      "step": 50
    },
    {
      "epoch": 0.7971014492753623,
      "grad_norm": 0.1591796875,
      "learning_rate": 2.4124187730720917e-05,
      "loss": 0.7908,
      "step": 55
    },
    {
      "epoch": 0.8695652173913043,
      "grad_norm": 0.17578125,
      "learning_rate": 1.0219546042925843e-05,
      "loss": 0.77,
      "step": 60
    },
    {
      "epoch": 0.9420289855072463,
      "grad_norm": 0.16796875,
      "learning_rate": 2.0470058747505516e-06,
      "loss": 0.7677,
      "step": 65
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.7884343266487122,
      "eval_runtime": 12.2005,
      "eval_samples_per_second": 4.918,
      "eval_steps_per_second": 0.656,
      "step": 69
    },
    {
      "epoch": 1.0,
      "step": 69,
      "total_flos": 5.119022417628365e+16,
      "train_loss": 0.934628030528193,
      "train_runtime": 403.9926,
      "train_samples_per_second": 1.366,
      "train_steps_per_second": 0.171
    }
  ],
  "logging_steps": 5,
  "max_steps": 69,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.119022417628365e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}