© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 59065
    Prev
    Next
    INDEX
    Explanations

    may

    np_max-act · gemini-2.0-flash

    The neuron fires on occurrences of the modal permission phrase “may” (as in “you may …”) typically found in Apache‐style license headers.

    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
    supplier
    -0.07
    чика
    -0.07
    ()})↵
    -0.06
     ils
    -0.06
    چی
    -0.06
    UnitTest
    -0.06
     типа
    -0.06
    azine
    -0.06
     WRONG
    -0.06
    County
    -0.06
    POSITIVE LOGITS
     kitab
    0.07
    ->{$
    0.07
    _markers
    0.07
     Ay
    0.06
    Sal
    0.06
    \Database
    0.06
     dismay
    0.06
     зай
    0.06
     Usually
    0.06
     electrons
    0.06
    Activations Density 0.005%

    No Known Activations