arXiv (All)
open access
Linear Separability of Activation Representations after Supervised Fine-Tuning on Incorrect Responses: A Study of Synthetic Dishonesty in Large Language Models
When a language model is fine-tuned to produce systematically incorrect responses, does this training leave a structured, linearly recoverable trace in its internal activations? We study this question in a controlled model-organism setting using five transformer architectures spanning 1.4 to 9 billi
Conceptio Open Archive
Analyze & explore related research in Conceptio
Connect this record directly into your research workflow or AI agent. Explore citations across open repositories, query full text, and export structured proof bundles.
This document is indexed with metadata only: full text is not available in the archive for this record. Open the official source ↗
Related documents
- From Detection to Refusal: Safer LLMs via Circuit-Guided Weight Scaling#819535
- From Detection to Refusal: Safer LLMs via Circuit-Guided Weight Scaling#809539
- How LLM Task-Adaptation Reshapes Alignment: A Multi-dimensional Study of Behavioral and Representational Drift#1129339
- Replay-Gated Neural Execution: Decoupling Persistent Behavioral Specifications from Neural Realizations in Frozen Language Models#1036926
- Where a Model Sends Its Own Repeated Token#1105739
Record · ID 973192
Retrieved via Conceptio: every document is proof-bundled with source, license, and retrieval metadata.
Retrieved via Conceptio: every document is proof-bundled with source, license, and retrieval metadata.
Bring this data into your agent: Connect via MCP or API · 20 free credits to start · no card required
Read the Docs → · View CLI source on GitHub · Conceptio Open Knowledge Archive
Read the Docs → · View CLI source on GitHub · Conceptio Open Knowledge Archive