© 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. 68886
    Prev
    Next
    INDEX
    Explanations

    CRUD

    np_max-act · gemini-2.0-flash

    The neuron flags user requests involving “CRUD” operations—i.e. mentions of building controllers or endpoints with Create/Read/Update/Delete functionality.

    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
     StringType
    -0.07
     mantle
    -0.06
    .WriteHeader
    -0.06
    -one
    -0.06
    asından
    -0.06
    -lnd
    -0.06
     відпов
    -0.06
     message
    -0.06
    Chair
    -0.06
    Ion
    -0.06
    POSITIVE LOGITS
    维
    0.07
    축
    0.07
    crud
    0.07
    Crud
    0.06
     ger
    0.06
    REV
    0.06
    τος
    0.06
    m
    0.06
    학년
    0.06
    tempts
    0.06
    Activations Density 0.003%

    No Known Activations