© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 38827
    Prev
    Next
    INDEX
    Explanations

    code translation/conversion

    np_max-act · gemini-2.0-flash

    The neuron fires on prose describing code‐conversion requests—especially phrases naming languages or functions (e.g. “version of …_encrypt,” “how to write … in Objective-C,” “I can find,” “in IOS”). It picks out instructions about translating or implementing code between programming languages.

    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
     diag
    -0.07
     Owl
    -0.07
     defeating
    -0.07
    DeltaTime
    -0.06
    .exam
    -0.06
    (source
    -0.06
    أ
    -0.06
     technicians
    -0.06
    .slot
    -0.06
    .return
    -0.06
    POSITIVE LOGITS
    스코
    0.06
    (fin
    0.06
    ='/
    0.06
     인기
    0.06
     spraw
    0.06
     getElement
    0.06
     vé
    0.06
     très
    0.06
     압
    0.06
     Unable
    0.06
    Activations Density 0.037%

    No Known Activations