© 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 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. 38883
    Prev
    Next
    INDEX
    Explanations

    ade

    np_max-act · gemini-2.0-flash

    The neuron detects tokens containing non-basic ASCII characters—i.e. letters with diacritics or in non-Latin scripts.

    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
    .ONE
    -0.07
    Free
    -0.07
    ètres
    -0.07
     patents
    -0.07
     bull
    -0.06
    CDF
    -0.06
    _null
    -0.06
    Phil
    -0.06
     corrected
    -0.06
     dns
    -0.06
    POSITIVE LOGITS
     شيء
    0.07
     Jak
    0.06
    自己的
    0.06
     kişilerin
    0.06
    тый
    0.06
     insanların
    0.06
    idUser
    0.06
     ItemType
    0.06
    implify
    0.06
     ακ
    0.06
    Activations Density 0.139%

    No Known Activations