mirror of
https://github.com/karpathy/nanochat.git
synced 2026-05-22 15:47:58 +00:00
The heuristic classifier was triggering web_search on 'who is your creator', 'who is manmohan sharma', 'who created you' etc — which returned irrelevant Tavily results (Tyler the Creator, Waaree CFO) when the model's SFT training already has the correct grounded identity answer. Added _IDENTITY_VETO_PATTERNS covering: self-referential questions, creator/maker/developer queries, competitor/provenance attacks (are you chatgpt/made by openai), samosaChaat/Manmohan name references, meta-questions (parameters/architecture/training/open source), greetings (hi/hello/hey), small talk, and writing/reasoning tasks that the model answers from memory. Veto runs before all positive classification. |
||
|---|---|---|
| .. | ||
| _model.py | ||
| _query_classifier.py | ||
| _tokenizer.py | ||
| _tools.py | ||
| serve.py | ||