© 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 EvalsExportsinterp-engineNEWAPI 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. 58971
    Prev
    Next
    INDEX
    Explanations

    Conversational documents

    np_max-act · gemini-2.0-flash

    This neuron activates on placeholder tokens used for anonymized or variable entity names (e.g. “NAME_1,” “NAME_2,” etc.).

    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
     вони
    -0.06
    _ma
    -0.06
     Abe
    -0.06
     restrictive
    -0.06
    _compile
    -0.06
     wooded
    -0.06
     ----------
    -0.06
    ['
    -0.06
     ⟨
    -0.06
    /ng
    -0.06
    POSITIVE LOGITS
     disrespect
    0.07
    @Entity
    0.07
    Runtime
    0.07
    .CASCADE
    0.07
     ErrorMessage
    0.06
     капіт
    0.06
     وج
    0.06
    	lcd
    0.06
    握
    0.06
    :variables
    0.06
    Activations Density 0.270%

    No Known Activations