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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 11-RESID-POST-AA
    11. 128407
    Prev
    Next
    INDEX
    Explanations

    question marks

    np_max-act · gemini-2.0-flash

    issues related to technical troubleshooting in computing or software.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron detects question boundaries, i.e. sentence-final question marks indicating a user’s question.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    instances of user questions (interrogative sentences / the question-mark token).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    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
     مدير
    -0.08
     �
    -0.07
     dismissed
    -0.07
    ρή
    -0.07
     معنی
    -0.07
     otherwise
    -0.06
     Expansion
    -0.06
     returning
    -0.06
     meio
    -0.06
     jiného
    -0.06
    POSITIVE LOGITS
    erts
    0.07
     갈
    0.06
    next
    0.06
    Ace
    0.06
    Contract
    0.06
     ragazzi
    0.06
     anticipating
    0.06
     fors
    0.06
    stp
    0.06
    elim
    0.06
    Activations Density 0.060%

    No Known Activations