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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 84707
    Prev
    Next
    INDEX
    Explanations

    hyphens

    np_max-act · gemini-2.0-flash

    The neuron activates on hyphen-prefixed segments of IUPAC-style chemical names (e.g. stereochemical or substituent descriptors like “-t”, “-hydro”, “-trifluoromethyl”).

    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
    An
    -0.07
     enchant
    -0.07
    -nine
    -0.06
    connexion
    -0.06
    (tol
    -0.06
    Ni
    -0.06
    For
    -0.06
     visually
    -0.06
     Campo
    -0.06
    IFICATION
    -0.06
    POSITIVE LOGITS
     foo
    0.06
    .displayName
    0.06
    وم
    0.06
    rax
    0.06
     FullName
    0.06
     апреля
    0.06
    ,Z
    0.06
     apar
    0.06
    _am
    0.06
     padding
    0.06
    Activations Density 0.035%

    No Known Activations