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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 91066
    Prev
    Next
    INDEX
    Explanations

    German language

    np_max-act · gemini-2.0-flash

    This neuron activates on German inflected descriptive words—especially adjectives and adverbs marked by suffixes like –lich, –ig, –isch, or –end.

    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
    원의
    -0.08
    three
    -0.07
     önceki
    -0.07
    buah
    -0.07
     THEIR
    -0.07
    ことを
    -0.07
     etme
    -0.07
     Něk
    -0.07
     Regiment
    -0.07
    oples
    -0.07
    POSITIVE LOGITS
    基本
    0.07
    _ai
    0.06
     chinese
    0.06
     OK
    0.06
     frankfurt
    0.06
     UP
    0.06
     haut
    0.06
    CONTENT
    0.06
    ISIBLE
    0.06
    續
    0.06
    Activations Density 0.066%

    No Known Activations