© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 129952
    Prev
    Next
    INDEX
    Explanations

    Nicknames/chat names

    np_max-act · gemini-2.0-flash

    The neuron activates on the dialogue’s speaker‐ID tokens (e.g. “NAME_1”, “NAME_2”, etc.), i.e. it is detecting turn‐labels for who is speaking.

    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
    _spin
    -0.07
     реч
    -0.06
     _______,
    -0.06
    ?>"><?
    -0.06
     همان
    -0.06
     Indianapolis
    -0.06
     sealing
    -0.06
    感情
    -0.06
    _tp
    -0.06
    すると
    -0.06
    POSITIVE LOGITS
     unsigned
    0.07
    enum
    0.07
    qm
    0.06
     عدد
    0.06
    \Model
    0.06
    yz
    0.06
    фи
    0.06
    فات
    0.06
     CALLBACK
    0.06
    이
    0.06
    Activations Density 0.018%

    No Known Activations