checkpoint-100/trainer_state.json

{
  "best_metric": 0.7295081967213115,
  "best_model_checkpoint": "./save_phi2_ft_lora/checkpoint-100",
  "epoch": 0.2857142857142857,
  "eval_steps": 100,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02857142857142857,
      "grad_norm": 0.18488438427448273,
      "learning_rate": 5e-06,
      "loss": 1.9926,
      "step": 10
    },
    {
      "epoch": 0.05714285714285714,
      "grad_norm": 0.20217478275299072,
      "learning_rate": 1e-05,
      "loss": 2.0428,
      "step": 20
    },
    {
      "epoch": 0.08571428571428572,
      "grad_norm": 0.22868461906909943,
      "learning_rate": 1.5e-05,
      "loss": 2.1981,
      "step": 30
    },
    {
      "epoch": 0.11428571428571428,
      "grad_norm": 0.31630581617355347,
      "learning_rate": 2e-05,
      "loss": 2.1773,
      "step": 40
    },
    {
      "epoch": 0.14285714285714285,
      "grad_norm": 0.46189257502555847,
      "learning_rate": 2.5e-05,
      "loss": 2.0399,
      "step": 50
    },
    {
      "epoch": 0.17142857142857143,
      "grad_norm": 1.0216264724731445,
      "learning_rate": 3e-05,
      "loss": 2.1152,
      "step": 60
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.6140756607055664,
      "learning_rate": 3.5e-05,
      "loss": 2.0064,
      "step": 70
    },
    {
      "epoch": 0.22857142857142856,
      "grad_norm": 0.7276609539985657,
      "learning_rate": 4e-05,
      "loss": 1.9443,
      "step": 80
    },
    {
      "epoch": 0.2571428571428571,
      "grad_norm": 0.694882869720459,
      "learning_rate": 4.5e-05,
      "loss": 1.9579,
      "step": 90
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 0.7517616748809814,
      "learning_rate": 5e-05,
      "loss": 1.9891,
      "step": 100
    },
    {
      "epoch": 0.2857142857142857,
      "eval_accuracy": 0.7295081967213115,
      "step": 100
    },
    {
      "epoch": 0.2857142857142857,
      "eval_loss": 1.9115574359893799,
      "eval_runtime": 333.2149,
      "eval_samples_per_second": 1.098,
      "eval_steps_per_second": 1.098,
      "step": 100
    }
  ],
  "logging_steps": 10,
  "max_steps": 700,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3147305934848e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}