trainer_state.json

{
  "best_metric": 0.7090692520141602,
  "best_model_checkpoint": "./phi3_4_finetuned/checkpoint-68",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 68,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2962962962962963,
      "grad_norm": 2.8733036518096924,
      "learning_rate": 0.0005882352941176471,
      "loss": 1.0773,
      "step": 10
    },
    {
      "epoch": 0.5925925925925926,
      "grad_norm": 0.5275980830192566,
      "learning_rate": 0.0009997733473639876,
      "loss": 0.9955,
      "step": 20
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 0.4500962793827057,
      "learning_rate": 0.0009957496810072027,
      "loss": 0.9132,
      "step": 30
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.7284564971923828,
      "eval_runtime": 12.4726,
      "eval_samples_per_second": 1.203,
      "eval_steps_per_second": 1.203,
      "step": 34
    },
    {
      "epoch": 1.1777777777777778,
      "grad_norm": 0.5402226448059082,
      "learning_rate": 0.0009867359188282193,
      "loss": 0.7281,
      "step": 40
    },
    {
      "epoch": 1.474074074074074,
      "grad_norm": 0.3571159839630127,
      "learning_rate": 0.0009728227911667932,
      "loss": 0.6595,
      "step": 50
    },
    {
      "epoch": 1.7703703703703704,
      "grad_norm": 0.4393090009689331,
      "learning_rate": 0.0009541503441850843,
      "loss": 0.6403,
      "step": 60
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.7090692520141602,
      "eval_runtime": 12.4743,
      "eval_samples_per_second": 1.202,
      "eval_steps_per_second": 1.202,
      "step": 68
    }
  ],
  "logging_steps": 10,
  "max_steps": 330,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "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": 2.486823536295936e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}