© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 59428
    Prev
    Next
    INDEX
    Explanations

    1 and 8

    np_max-act · gemini-2.0-flash

    This neuron detects numeric width specifiers in C/C++ bitfield declarations (the “:1”, “:2” etc. after field names).

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    	F
    -0.07
     Gold
    -0.06
    _dest
    -0.06
     Evo
    -0.06
    webView
    -0.06
    Maps
    -0.06
    $current
    -0.06
     Rockets
    -0.06
     Minutes
    -0.06
    [];
↵
    -0.06
    POSITIVE LOGITS
    ồn
    0.07
    _episodes
    0.07
    ю
    0.07
     phái
    0.07
    (dataset
    0.06
    _decl
    0.06
    śmy
    0.06
    ，还
    0.06
    ulado
    0.06
    _inode
    0.06
    Activations Density 0.002%

    No Known Activations