© 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 (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 25356
    Prev
    Next
    INDEX
    Explanations

    Medical research

    np_max-act · gemini-2.0-flash

    The neuron fires on scientific “meta‐language” signaling research implications or biological mechanisms (e.g. terms indicating susceptibility, “potential” candidates, providing “clues,” and “pathogenesis”).

    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
    .WebServlet
    -0.06
     skate
    -0.06
     ков
    -0.06
     Scene
    -0.06
    Gary
    -0.06
    ír
    -0.06
     Joanna
    -0.06
     opera
    -0.06
    áže
    -0.06
     być
    -0.06
    POSITIVE LOGITS
    ());//
    0.07
     trays
    0.07
    té
    0.06
    ара
    0.06
     акку
    0.06
     nombres
    0.06
    _BLE
    0.06
    RET
    0.06
     후
    0.06
     heaps
    0.06
    Activations Density 0.041%

    No Known Activations