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

    Neuronpedia

    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. 9132
    Prev
    Next
    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    The neuron consistently lights up on occurrences of the “fd5” prefix (e.g. in header‐include paths and identifiers like fd5_context), i.e. it detects tokens containing “fd5_.”

    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
     copyrighted
    -0.07
    classname
    -0.06
    Dispatch
    -0.06
     viewpoint
    -0.06
    _primitive
    -0.06
    >=
    -0.06
     marshaller
    -0.06
     правил
    -0.06
    getNext
    -0.06
     入
    -0.06
    POSITIVE LOGITS
    /thumb
    0.07
     TreeSet
    0.07
     Catalonia
    0.07
    लग
    0.07
    姉
    0.07
     DEBUG
    0.07
     Мал
    0.07
    玄
    0.06
     Croatian
    0.06
    .command
    0.06
    Activations Density 0.221%

    No Known Activations