{
  "path": "artifacts/bigram/llada.pt",
  "checkpoint_version": 2,
  "successor_tokens": 76084287,
  "successor_marginal": "additive-beta-smoothed successor counts from observed pairs",
  "observed_pairs": 5109347,
  "tokens": 77064287,
  "tokenizer": "base_models/models--GSAI-ML--LLaDA-8B-Instruct/snapshots/08b83a6feb34df1a6011b80c3c00c7563e963b07",
  "tokenizer_sha256": "2d2bab73abe3bf56a7f5d4572eee57e7705387526c05f043b8412a86b6175022",
  "source": "data/lad-training-1m-256-long",
  "split": "train",
  "max_documents": null,
  "text_column": "output",
  "response_eos": true,
  "documents": 980000,
  "model_vocab_size": 126464,
  "tokenizer_length": 126349,
  "tokenizer_model_id": "GSAI-ML/LLaDA-8B-Instruct",
  "model_revision": "08b83a6feb34df1a6011b80c3c00c7563e963b07"
}