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. 85945
    Prev
    Next
    INDEX
    Explanations

    ://

    np_max-act · gemini-2.0-flash

    This neuron activates on parts of URLs (protocol separators like “://” and nearby URL path/domain tokens).

    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
    ِ
    -0.07
    �
    -0.06
    pike
    -0.06
     phản
    -0.06
    cut
    -0.06
     heater
    -0.06
    �
    -0.06
    'S
    -0.06
     Vulcan
    -0.06
    誌
    -0.06
    POSITIVE LOGITS
     )]↵
    0.08
     info
    0.07
    /auth
    0.07
    =config
    0.07
    _Move
    0.07
    .GetKey
    0.07
    ============↵
    0.07
     Miche
    0.07
    .aut
    0.07
     getKey
    0.06
    Activations Density 0.016%

    No Known Activations