© 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. 11-RESID-POST-AA
    11. 102173
    Prev
    Next
    INDEX
    Explanations

    code or documentation

    np_max-act · gemini-2.0-flash

    technical jargon and keywords related to software development and coding practices.

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

    This neuron fires on common English “function” words (the, and, to, in, if, this, etc.) appearing in code comments or documentation.

    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
    unnable
    -0.07
    稱
    -0.06
    ()).
    -0.06
    {T
    -0.06
     certains
    -0.06
     (>
    -0.06
     mastering
    -0.06
    プレ
    -0.06
     Bowie
    -0.06
     nomin
    -0.06
    POSITIVE LOGITS
     bắc
    0.06
     δημο
    0.06
    egral
    0.06
     hoax
    0.06
     prayers
    0.06
    okia
    0.06
     rept
    0.06
     eig
    0.06
    _DUMP
    0.06
    ución
    0.06
    Activations Density 0.043%

    No Known Activations