© 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. 58284
    Prev
    Next
    INDEX
    Explanations

    Scientific abstracts

    np_max-act · gemini-2.0-flash

    This neuron fires on numeric expressions and enumerations (counts, digits, numbered compound listings) in scientific descriptions.

    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
     Piece
    -0.07
    	short
    -0.06
    (lr
    -0.06
    (Input
    -0.06
    SqlCommand
    -0.06
    <header
    -0.06
    	act
    -0.06
    failure
    -0.06
    	animation
    -0.06
    experimental
    -0.06
    POSITIVE LOGITS
    Independ
    0.07
     جزئ
    0.06
     campaign
    0.06
     žád
    0.06
    CEED
    0.06
     Television
    0.06
    _UD
    0.06
     coun
    0.06
     Qué
    0.06
     През
    0.05
    Activations Density 0.013%

    No Known Activations