© 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. 15-RESID-POST-AA
    11. 18904
    Prev
    Next
    INDEX
    Explanations

    Question/answer forum

    np_max-act · gemini-2.0-flash

    tokens that mark or begin the assistant's reply (assistant-role / reply-start tokens).

    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_15/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    128
    -0.07
    аче
    -0.07
    _show
    -0.07
    :semicolon
    -0.07
    ,h
    -0.06
    Recursive
    -0.06
     '!
    -0.06
     Frequently
    -0.06
    IPC
    -0.06
    irus
    -0.06
    POSITIVE LOGITS
    <|eot_id|>
    0.06
     perv
    0.06
    -model
    0.06
     elabor
    0.06
     anda
    0.06
     quiere
    0.06
     veel
    0.06
    jac
    0.06
    Sele
    0.06
     nhẹ
    0.06
    Activations Density 0.026%

    No Known Activations