© 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. 74785
    Prev
    Next
    INDEX
    Explanations

    code fragments

    np_max-act · gemini-2.0-flash

    The neuron detects the assistant’s refusal or apology statements when declining inappropriate or disallowed requests.

    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
     unresolved
    -0.07
    服务器
    -0.06
    оряд
    -0.06
    一度
    -0.06
    _LCD
    -0.06
     effortlessly
    -0.06
     stripes
    -0.06
     xung
    -0.06
    (src
    -0.06
    ublik
    -0.06
    POSITIVE LOGITS
    cantidad
    0.07
    .Floor
    0.07
     getP
    0.06
     Crush
    0.06
    ').'</
    0.06
    ルの
    0.06
    еления
    0.06
    .Override
    0.06
    alse
    0.06
     heal
    0.06
    Activations Density 0.027%

    No Known Activations