© 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
    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. 30107
    Prev
    Next
    INDEX
    Explanations

    the

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on tokens that are characteristic of Portuguese text (e.g. “um,” “explorador,” “italiano,” inflected endings like “-adores,” accented forms), so it’s detecting Portuguese-language words.

    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
     Guild
    -0.06
    250
    -0.06
    po
    -0.06
     лож
    -0.06
     Testament
    -0.06
     Esper
    -0.06
     coercion
    -0.06
     logical
    -0.06
    CppTypeDefinition
    -0.06
    igated
    -0.06
    POSITIVE LOGITS
    som
    0.07
    -piece
    0.07
    (">
    0.06
    âu
    0.06
     MISSING
    0.06
    .Local
    0.06
    (LL
    0.06
     Qty
    0.06
    handle
    0.06
     ```
    0.06
    Activations Density 0.198%

    No Known Activations