© 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. 91343
    Prev
    Next
    INDEX
    Explanations

    to

    np_max-act · gemini-2.0-flash

    This neuron detects the patent-style objective or purpose statements that introduce what the invention “provides” or “is an object of.”

    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
    аки
    -0.08
     habitual
    -0.07
    mouseover
    -0.06
    About
    -0.06
    uParam
    -0.06
    逃
    -0.06
    INCT
    -0.06
     Bust
    -0.06
     survives
    -0.06
    .colors
    -0.06
    POSITIVE LOGITS
    dad
    0.07
     aut
    0.07
    ']).
    0.06
    -option
    0.06
     lambda
    0.06
     Windows
    0.06
     öff
    0.06
    	field
    0.06
     ниж
    0.06
     Mi
    0.06
    Activations Density 0.006%

    No Known Activations