© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 92009
    Prev
    Next
    INDEX
    Explanations

    data information issues

    np_max-act · gemini-2.0-flash

    The neuron activates on words and phrases about factual accuracy and consistency—e.g. “factually,” “misrepresent,” “mistakes”—so it’s detecting discussion of fact‐checking or factual consistency.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     donors
    -0.07
     Pagination
    -0.07
    pr
    -0.07
    kiye
    -0.07
    İs
    -0.06
    .HttpServlet
    -0.06
     Trit
    -0.06
    encias
    -0.06
    .Operation
    -0.06
     Verse
    -0.06
    POSITIVE LOGITS
    _sy
    0.07
     گوش
    0.07
    털
    0.07
     آقای
    0.07
    (DATA
    0.06
    گونه
    0.06
     repmat
    0.06
    incre
    0.06
     mutableListOf
    0.06
    _procs
    0.06
    Activations Density 0.001%

    No Known Activations