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. 84461
    Prev
    Next
    INDEX
    Explanations

    scientific data

    np_max-act · gemini-2.0-flash

    This neuron fires on alphanumeric drug or compound identifiers formatted as names with hyphens and numbers.

    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
     Router
    -0.07
    Place
    -0.07
     เท
    -0.06
    _Util
    -0.06
    err
    -0.06
    .popup
    -0.06
    .invalidate
    -0.06
    ugas
    -0.06
    uplicate
    -0.06
    "As
    -0.06
    POSITIVE LOGITS
    法
    0.06
    キング
    0.06
     procedural
    0.06
    .tableLayoutPanel
    0.06
     vrai
    0.06
     캐
    0.06
     vfs
    0.06
    ؤول
    0.06
    비아
    0.06
     мест
    0.05
    Activations Density 0.007%

    No Known Activations