trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.876543209876543,
  "eval_steps": 500,
  "global_step": 1400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.7054673721340388,
      "grad_norm": 0.20514033734798431,
      "learning_rate": 0.0002,
      "loss": 1.3614,
      "step": 100
    },
    {
      "epoch": 1.4109347442680775,
      "grad_norm": 0.18633601069450378,
      "learning_rate": 0.0002,
      "loss": 1.2692,
      "step": 200
    },
    {
      "epoch": 2.1164021164021163,
      "grad_norm": 0.22059209644794464,
      "learning_rate": 0.0002,
      "loss": 1.2149,
      "step": 300
    },
    {
      "epoch": 2.821869488536155,
      "grad_norm": 0.2522888779640198,
      "learning_rate": 0.0002,
      "loss": 1.1426,
      "step": 400
    },
    {
      "epoch": 3.527336860670194,
      "grad_norm": 0.3600117266178131,
      "learning_rate": 0.0002,
      "loss": 1.0687,
      "step": 500
    },
    {
      "epoch": 4.232804232804233,
      "grad_norm": 0.617989718914032,
      "learning_rate": 0.0002,
      "loss": 0.9957,
      "step": 600
    },
    {
      "epoch": 4.938271604938271,
      "grad_norm": 0.5240411162376404,
      "learning_rate": 0.0002,
      "loss": 0.973,
      "step": 700
    },
    {
      "epoch": 5.64373897707231,
      "grad_norm": 0.615224301815033,
      "learning_rate": 0.0002,
      "loss": 0.8472,
      "step": 800
    },
    {
      "epoch": 6.349206349206349,
      "grad_norm": 0.6250160932540894,
      "learning_rate": 0.0002,
      "loss": 0.7726,
      "step": 900
    },
    {
      "epoch": 7.054673721340388,
      "grad_norm": 0.7413871884346008,
      "learning_rate": 0.0002,
      "loss": 0.7079,
      "step": 1000
    },
    {
      "epoch": 7.760141093474427,
      "grad_norm": 1.084841012954712,
      "learning_rate": 0.0002,
      "loss": 0.6037,
      "step": 1100
    },
    {
      "epoch": 8.465608465608465,
      "grad_norm": 1.128200650215149,
      "learning_rate": 0.0002,
      "loss": 0.5464,
      "step": 1200
    },
    {
      "epoch": 9.171075837742505,
      "grad_norm": 1.0372769832611084,
      "learning_rate": 0.0002,
      "loss": 0.5376,
      "step": 1300
    },
    {
      "epoch": 9.876543209876543,
      "grad_norm": 1.0853991508483887,
      "learning_rate": 0.0002,
      "loss": 0.475,
      "step": 1400
    }
  ],
  "logging_steps": 100,
  "max_steps": 1410,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 100,
  "total_flos": 1.646216165505024e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}