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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 57029
    Prev
    Next
    INDEX
    Explanations

    commas, parentheses

    np_max-act · gemini-2.0-flash

    This neuron detects longer, Latinate words—especially nouns and adjectives formed with suffixes like ‑ion, ‑ment, ‑ive, or ‑ist.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    clause- and list-separating punctuation, especially commas and closing parentheses in enumerations or parenthetical asides.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    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
     deserted
    -0.07
     Bug
    -0.07
    _slave
    -0.07
     unused
    -0.06
     centroid
    -0.06
     THE
    -0.06
    avia
    -0.06
    ForRow
    -0.06
     ARG
    -0.06
     the
    -0.06
    POSITIVE LOGITS
     n
    0.07
    방법
    0.06
    .'.$
    0.06
     با
    0.06
    	virtual
    0.06
    με
    0.06
    альні
    0.06
     RM
    0.06
     ${(
    0.06
     in
    0.06
    Activations Density 0.252%

    No Known Activations