Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 40595
    Prev
    Next
    INDEX
    Explanations

    forum posts

    np_max-act · gemini-2.0-flash

    This neuron detects Spanish-language assistance or help‐offer phrases (e.g. “¿En qué puedo ayudarte?”, “Estoy aquí para ayudar”, “¿Hay algo…?”).

    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
     servers
    -0.07
    .contacts
    -0.07
    aver
    -0.07
     Dante
    -0.07
    迹
    -0.06
     today
    -0.06
    obe
    -0.06
     Stamp
    -0.06
    طي
    -0.06
    _tmp
    -0.06
    POSITIVE LOGITS
     Xã
    0.07
    Middleware
    0.06
    해야
    0.06
    镇
    0.06
    erli
    0.06
     loung
    0.06
     bolster
    0.06
    Undefined
    0.06
    .AdapterView
    0.06
     automobiles
    0.06
    Activations Density 0.070%

    No Known Activations