© 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. 55924
    Prev
    Next
    INDEX
    Explanations

    physics and mathematics

    np_max-act · gemini-2.0-flash

    This neuron spotlights technical terms containing the substrings “free” (as in “quasi-free,” “free risk,” etc.) or “mutation” (as in “commutation,” “anticommutation,” 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
    ć
    -0.07
     jed
    -0.06
    ỉnh
    -0.06
     cafe
    -0.06
     café
    -0.06
    τερο
    -0.06
     seizure
    -0.06
     unveiled
    -0.06
    _inverse
    -0.06
     Pr
    -0.06
    POSITIVE LOGITS
    _COM
    0.07
    _typeof
    0.07
    지를
    0.07
    Net
    0.06
     Peng
    0.06
     Yorkshire
    0.06
    _LAST
    0.06
    FACT
    0.06
     McGr
    0.06
     Svens
    0.06
    Activations Density 0.081%

    No Known Activations