Base GPT-2 XL Better SFT — HF-Only Assistant Coherence Continuation
Continuation checkpoint trained from Darkstorm1826/base-gpt2xl-better-sft-experimental using a JAX/Optax TPU full fine-tune pass.
This experimental continuation uses /kaggle/working/distilled.jsonl, built only from Hugging Face datasets.
Focus:
- broader assistant behavior
- chat/instruction following
- creative writing
- everyday procedural answers
- context-grounded QA
- truthfulness and hallucination reduction
- cleaner stopping behavior
Prompt format:
<|system|> You are a helpful, honest, and careful AI assistant.<|end|> <|user|> Question here<|end|> <|assistant|> Answer here<|end|>
- Downloads last month
- 38
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for Darkstorm1826/base-gpt2xl-better-sft-experimental
Unable to build the model tree, the base model loops to the model itself. Learn more.