© 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
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 8988
    Prev
    Next
    INDEX
    Explanations

    to

    np_max-act · gemini-2.0-flash

    The neuron activates on explanatory phrases in code commentary—especially tokens of the form “the number to be …” that describe a function’s parameter.

    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
     поход
    -0.07
    iona
    -0.07
    395
    -0.06
    �
    -0.06
     grandma
    -0.06
    考试
    -0.06
    Supplier
    -0.06
     afflicted
    -0.06
     Aff
    -0.06
    FULL
    -0.06
    POSITIVE LOGITS
     gilt
    0.06
     twe
    0.06
    {|
    0.06
     dramatically
    0.06
    ْل
    0.06
    .titleLabel
    0.06
    ="[
    0.06
    .friend
    0.06
    allel
    0.06
     sonic
    0.06
    Activations Density 0.056%

    No Known Activations