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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 15-RESID-POST-AA
    11. 34111
    Prev
    Next
    INDEX
    Explanations

    yes

    np_max-act · gemini-2.0-flash

    Cyrillic а

    np_max-act · gpt-5-miniTriggered by @yooniel31

    the Russian affirmative "да" (the word "yes"), especially when starting a response.

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_15/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    isinin
    -0.07
    lesen
    -0.07
     DB
    -0.06
    кових
    -0.06
     Folding
    -0.06
     propagated
    -0.06
    *S
    -0.06
     folding
    -0.06
     Economy
    -0.06
     пас
    -0.06
    POSITIVE LOGITS
     Bund
    0.07
    UNC
    0.07
     гид
    0.06
    レビ
    0.06
     boolean
    0.06
     Failed
    0.06
     Win
    0.06
     formData
    0.06
    _he
    0.06
    驗
    0.06
    Activations Density 0.030%

    No Known Activations