checkpoint-404/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 404,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.25,
      "grad_norm": 0.4402466118335724,
      "learning_rate": 0.00019876883405951377,
      "loss": 1.5784,
      "step": 50
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5029518604278564,
      "learning_rate": 0.00019510565162951537,
      "loss": 1.4446,
      "step": 100
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5096282958984375,
      "learning_rate": 0.0001891006524188368,
      "loss": 1.4047,
      "step": 150
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.508622407913208,
      "learning_rate": 0.00018090169943749476,
      "loss": 1.3806,
      "step": 200
    },
    {
      "epoch": 1.24,
      "grad_norm": 0.6428921222686768,
      "learning_rate": 0.00017071067811865476,
      "loss": 1.2506,
      "step": 250
    },
    {
      "epoch": 1.49,
      "grad_norm": 0.6709182262420654,
      "learning_rate": 0.00015877852522924732,
      "loss": 1.254,
      "step": 300
    },
    {
      "epoch": 1.73,
      "grad_norm": 0.6514674425125122,
      "learning_rate": 0.00014539904997395468,
      "loss": 1.2417,
      "step": 350
    },
    {
      "epoch": 1.98,
      "grad_norm": 0.7174981236457825,
      "learning_rate": 0.00013090169943749476,
      "loss": 1.252,
      "step": 400
    }
  ],
  "logging_steps": 50,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 5255149859143680.0,
  "train_batch_size": 7,
  "trial_name": null,
  "trial_params": null
}