© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 109886
    Prev
    Next
    INDEX
    Explanations

    email

    np_max-act · gemini-2.0-flash

    This neuron fires on words naming communication or embedding channels (e.g. “email,” “IM,” “document,” “website”) within link‐sharing instructions.

    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
     उठ
    -0.07
    uset
    -0.06
     облі
    -0.06
     destabil
    -0.06
    _CONTAINER
    -0.06
     Роз
    -0.06
    .quest
    -0.06
     реп
    -0.06
    ê
    -0.06
    _PREFIX
    -0.06
    POSITIVE LOGITS
    Marks
    0.07
     leaked
    0.06
    炸
    0.06
     charismatic
    0.06
     simplify
    0.06
    documento
    0.06
    .ctrl
    0.06
    'H
    0.06
     glyph
    0.06
     blasts
    0.06
    Activations Density 0.000%

    No Known Activations