© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 129810
    Prev
    Next
    INDEX
    Explanations

    1

    np_max-act · gemini-2.0-flash

    This neuron detects the numbered “Specific Aims” list markers (the digits 1, 2, 3, etc., often with surrounding colons or parentheses) in research proposal text.

    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
    amaz
    -0.07
     switch
    -0.07
    Requested
    -0.07
     праців
    -0.07
     لإ
    -0.07
    -demand
    -0.06
     required
    -0.06
     Regel
    -0.06
    ality
    -0.06
     sing
    -0.06
    POSITIVE LOGITS
     reminiscent
    0.07
     restrain
    0.06
     BUSINESS
    0.06
    untas
    0.06
     bum
    0.06
     korum
    0.06
    ’une
    0.06
     cols
    0.06
    document
    0.06
    .ColumnName
    0.06
    Activations Density 0.010%

    No Known Activations