checkpoint-69/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 69,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3,
      "grad_norm": 0.621504545211792,
      "learning_rate": 3e-05,
      "loss": 2.5074,
      "step": 7
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.4357547461986542,
      "learning_rate": 2.6612903225806453e-05,
      "loss": 2.4362,
      "step": 14
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5074889063835144,
      "learning_rate": 2.3225806451612902e-05,
      "loss": 2.5027,
      "step": 21
    },
    {
      "epoch": 1.22,
      "grad_norm": 0.6310410499572754,
      "learning_rate": 1.9838709677419355e-05,
      "loss": 2.5137,
      "step": 28
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.5066408514976501,
      "learning_rate": 1.6451612903225807e-05,
      "loss": 2.5501,
      "step": 35
    },
    {
      "epoch": 1.83,
      "grad_norm": 0.5650488138198853,
      "learning_rate": 1.3064516129032258e-05,
      "loss": 2.1844,
      "step": 42
    },
    {
      "epoch": 2.13,
      "grad_norm": 0.44675177335739136,
      "learning_rate": 9.67741935483871e-06,
      "loss": 2.3875,
      "step": 49
    },
    {
      "epoch": 2.43,
      "grad_norm": 0.44878822565078735,
      "learning_rate": 6.290322580645162e-06,
      "loss": 2.4376,
      "step": 56
    },
    {
      "epoch": 2.74,
      "grad_norm": 0.3949732780456543,
      "learning_rate": 2.9032258064516128e-06,
      "loss": 2.3816,
      "step": 63
    }
  ],
  "logging_steps": 7,
  "max_steps": 69,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 148233963700224.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}