© 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. 56547
    Prev
    Next
    INDEX
    Explanations

    defunct

    np_max-act · gemini-2.0-flash

    This neuron activates on the “Def” (short for “Defunct”) label in category tags, i.e. it detects when an article is marked as defunct.

    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
     пред
    -0.06
    Bob
    -0.06
     Reid
    -0.06
    boru
    -0.06
     ridiculously
    -0.06
    丘
    -0.06
     Beng
    -0.06
     bounding
    -0.06
    Brandon
    -0.06
     về
    -0.06
    POSITIVE LOGITS
     силу
    0.07
     statement
    0.06
     стать
    0.06
    Searching
    0.06
    ΑΛ
    0.06
     statements
    0.06
    _family
    0.06
     Pearl
    0.06
    _lex
    0.06
    cano
    0.06
    Activations Density 0.006%

    No Known Activations