© 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
    4. Resid Post - 131k
    5. 7-RESID-POST-AA
    6. 121019
    Prev
    Next
    INDEX
    Explanations

    {

    np_max-act · gemini-2.0-flash

    the neuron detects LaTeX/math-mode content such as formula tokens, math environments, and equation markup.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_7/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.7.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    确认
    -0.07
    лик
    -0.06
     пла
    -0.06
    ldb
    -0.06
    POSITORY
    -0.06
    ,q
    -0.06
     accel
    -0.06
     redefine
    -0.06
     unearth
    -0.06
    かけ
    -0.06
    POSITIVE LOGITS
     antes
    0.07
     Astros
    0.07
     beef
    0.06
    .drawRect
    0.06
    áu
    0.06
    _ACTIVITY
    0.06
    agon
    0.06
     toute
    0.06
    uffling
    0.06
     vie
    0.06
    Activations Density 0.019%

    No Known Activations