© 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. 6532
    Prev
    Next
    INDEX
    Explanations

    of

    np_max-act · gemini-2.0-flash

    the neuron activates on words that denote legal structural references (e.g. “section,” “subdivision,” “division,” “portions”).

    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
     invested
    -0.07
     人
    -0.06
    cház
    -0.06
    ата
    -0.06
    xffff
    -0.06
     winter
    -0.06
    DataManager
    -0.06
     occurred
    -0.06
     stress
    -0.06
    tons
    -0.06
    POSITIVE LOGITS
    lineno
    0.07
    .setHeader
    0.06
     Định
    0.06
     mainAxisAlignment
    0.06
     Exceptions
    0.06
     http
    0.06
     gratuiti
    0.06
     Fast
    0.06
     Wishlist
    0.06
    TAIL
    0.06
    Activations Density 0.117%

    No Known Activations