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. 86002
    Prev
    Next
    INDEX
    Explanations

    HTML code

    np_max-act · gemini-2.0-flash

    The neuron fires on Sass/SCSS parent‐selector interpolation syntax (the “#{&}” pattern).

    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
     dicks
    -0.07
     Joined
    -0.07
    .Amount
    -0.06
    815
    -0.06
     emotional
    -0.06
     IH
    -0.06
     >>↵↵
    -0.06
    _SHARE
    -0.06
     níž
    -0.06
    _watch
    -0.06
    POSITIVE LOGITS
    ==="
    0.07
    ạt
    0.06
     Texture
    0.06
     어머니
    0.06
    Monster
    0.06
    vably
    0.06
     padded
    0.06
     emission
    0.06
    /",
    0.06
    のお
    0.06
    Activations Density 0.100%

    No Known Activations