© 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. 44194
    Prev
    Next
    INDEX
    Explanations

    code, programming

    np_max-act · gemini-2.0-flash

    This neuron detects code-like member declarations, especially words such as “method” or “property” in programming contexts.

    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
    주시
    -0.07
    something
    -0.07
    unwrap
    -0.07
     beating
    -0.07
     cadre
    -0.07
    eco
    -0.07
    ับความ
    -0.07
     prisons
    -0.07
    riad
    -0.07
     tries
    -0.07
    POSITIVE LOGITS
    中文
    0.06
     inclined
    0.06
     *,
    0.06
    SerializedName
    0.06
     FAG
    0.06
    _PAR
    0.06
    _ASSOC
    0.06
     waren
    0.05
     itk
    0.05
    \\.
    0.05
    Activations Density 0.048%

    No Known Activations