© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 68501
    Prev
    Next
    INDEX
    Explanations

    Sports

    np_max-act · gemini-2.0-flash

    This neuron detects phrases of the form “on the <playing surface>,” i.e. “on the court,” “on the ice,” “on the field,” “on the pitch,” etc.

    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
     demok
    -0.06
    _npc
    -0.06
    =args
    -0.06
    ussia
    -0.06
    EPROM
    -0.06
     itemName
    -0.06
    MEM
    -0.06
    _replace
    -0.06
    //------------------------------------------------
    -0.05
    (nodes
    -0.05
    POSITIVE LOGITS
     Tee
    0.07
     Cognitive
    0.06
    측
    0.06
    /select
    0.06
    licit
    0.06
    metics
    0.06
    -nil
    0.06
    Fitness
    0.06
     Royale
    0.06
    uri
    0.06
    Activations Density 0.010%

    No Known Activations