© 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. 29914
    Prev
    Next
    INDEX
    Explanations

    strings

    np_max-act · gemini-2.0-flash

    The neuron fires on tokens from string‐handling calls (e.g. methods like Split, Fields, Contains, Trim, Index, LastIndex, HasPrefix, EqualFold, etc.) indicating spots where string‐manipulation functions are invoked.

    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
     중국
    -0.07
    ující
    -0.06
    ["$
    -0.06
     settle
    -0.06
     Kunden
    -0.06
     secretion
    -0.06
     yön
    -0.06
     penalty
    -0.06
    876
    -0.06
     highest
    -0.06
    POSITIVE LOGITS
    airy
    0.08
     παρα
    0.06
     Ã
    0.06
     disb
    0.06
     ab
    0.06
     rb
    0.06
     Todos
    0.06
     checkBox
    0.06
     getName
    0.06
     fright
    0.06
    Activations Density 0.003%

    No Known Activations