© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 128849
    Prev
    Next
    INDEX
    Explanations

    URLs and HTML

    np_max-act · gemini-2.0-flash

    The neuron is triggered by the named-capture group syntax in Django URL regexes (the “(?P<name>…)” parts defining URL parameters).

    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
    elage
    -0.07
    Tx
    -0.06
    zim
    -0.06
    pricing
    -0.06
    EMENT
    -0.06
     chóng
    -0.06
     Shopify
    -0.06
    وران
    -0.06
     entreprene
    -0.05
    714
    -0.05
    POSITIVE LOGITS
    ตรว
    0.08
    .Sign
    0.07
    ミュ
    0.06
     anarchists
    0.06
     тяжел
    0.06
     members
    0.06
     reopening
    0.06
     digit
    0.06
    (an
    0.06
     Attribute
    0.06
    Activations Density 0.002%

    No Known Activations