© 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. 7-RESID-POST-AA
    11. 78100
    Prev
    Next
    INDEX
    Explanations

    \Desktop

    np_max-act · gemini-2.0-flash

    Windows-style file paths and filenames (drive letters with backslashes and path components).

    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_7/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.7.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     synthes
    -0.07
    cursor
    -0.07
     ","
    -0.06
     Nor
    -0.06
    Cast
    -0.06
     SIZE
    -0.06
     SAN
    -0.06
    .photos
    -0.06
    ixels
    -0.06
     علی
    -0.06
    POSITIVE LOGITS
    /Desktop
    0.10
    \Desktop
    0.09
     chiar
    0.07
    DSA
    0.07
    拒
    0.07
    ATION
    0.06
    	anim
    0.06
     classic
    0.06
     derec
    0.06
    ']]
    0.06
    Activations Density 0.001%

    No Known Activations