Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 130182
    Prev
    Next
    INDEX
    Explanations

    computer code/database

    np_max-act · gemini-2.0-flash

    This neuron detects the substring “cp” when it appears as part of database identifiers (table or column names) in SQL code.

    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
    ,如果
    -0.07
    .BAD
    -0.06
    ों
    -0.06
     Valle
    -0.06
     вз
    -0.06
     Ji
    -0.06
    舒
    -0.06
     disappe
    -0.06
    垂
    -0.05
    ้าช
    -0.05
    POSITIVE LOGITS
     Faces
    0.07
    .button
    0.07
    _comb
    0.07
    (","
    0.06
     घटन
    0.06
    .onClick
    0.06
     Retrieves
    0.06
     userService
    0.06
    	bg
    0.06
     sino
    0.06
    Activations Density 0.056%

    No Known Activations