© 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. 13516
    Prev
    Next
    INDEX
    Explanations

    int

    np_max-act · gemini-2.0-flash

    This neuron responds to C-language type and storage-class keywords (e.g. “static,” “struct,” “int,” “void”) typically used in declarations.

    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
     uzav
    -0.06
     ca
    -0.06
     nie
    -0.06
     Press
    -0.06
     Summon
    -0.06
     offs
    -0.06
    (!$
    -0.06
     FAIL
    -0.06
    _voice
    -0.06
    ências
    -0.06
    POSITIVE LOGITS
    φορ
    0.07
     tribes
    0.07
     кл
    0.06
     metodo
    0.06
    ']])↵
    0.06
     Κ
    0.06
     retro
    0.06
    CGRect
    0.06
    ภาพยนตร
    0.06
    �력
    0.06
    Activations Density 0.002%

    No Known Activations