Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 151
    Prev
    Next
    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash

    The neuron fires on function words and auxiliaries that kick off new sentences or clauses—i.e. discourse-marker tokens like “For,” “As,” “If,” “When,” “can,” “They,” “does,” 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
     plants
    -0.07
     dependencies
    -0.07
     chịu
    -0.07
     Seeing
    -0.07
     modulation
    -0.06
    .WebServlet
    -0.06
     residing
    -0.06
    алося
    -0.06
     blue
    -0.06
    jets
    -0.06
    POSITIVE LOGITS
    ΟΓ
    0.07
     Geological
    0.06
    ormsg
    0.06
    ỳ
    0.06
    ्मक
    0.06
     sociology
    0.06
    /year
    0.06
     Comment
    0.06
    le
    0.06
     Timeline
    0.06
    Activations Density 0.017%

    No Known Activations