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

    Neuronpedia

    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. 108899
    Prev
    Next
    INDEX
    Explanations

    legal intervention

    np_max-act · gemini-2.0-flash

    This neuron detects occurrences of “intervene” and its morphological variants (e.g. intervention, intervenor).

    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
     jylland
    -0.07
    Bone
    -0.06
     coerc
    -0.06
     Lean
    -0.06
     closure
    -0.06
     CELL
    -0.06
     respectively
    -0.06
    cursor
    -0.06
    _mapping
    -0.06
     coercion
    -0.06
    POSITIVE LOGITS
    isha
    0.07
    .Sequential
    0.07
    (['/
    0.06
     clearColor
    0.06
    (Runtime
    0.06
    .articles
    0.06
    grounds
    0.06
    <typeof
    0.06
     جع
    0.06
     urg
    0.06
    Activations Density 0.002%

    No Known Activations