© 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. 101166
    Prev
    Next
    INDEX
    Explanations

    ly

    np_max-act · gemini-2.0-flash

    language that expresses spatial direction or positional relationships, especially directional prepositions/adverbs indicating movement or location.

    oai_token-act-pair · gpt-5
    Triggered by @vetterc0
    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
     hypothesis
    -0.07
     stochastic
    -0.07
     finite
    -0.07
     farewell
    -0.07
     touring
    -0.07
    ASURE
    -0.06
    .want
    -0.06
     Linden
    -0.06
                     
    -0.06
     SHALL
    -0.06
    POSITIVE LOGITS
    \Action
    0.07
     بي
    0.06
    ROOT
    0.06
    acf
    0.06
    níku
    0.06
    ename
    0.06
    _services
    0.06
    .nodeType
    0.06
     рік
    0.06
    Open
    0.05
    Activations Density 1.260%

    No Known Activations