© 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. 58634
    Prev
    Next
    INDEX
    Explanations

    distances

    np_max-act · gemini-2.0-flash

    The neuron fires on decimal number tokens inside parentheses that represent unit‐conversion distances (e.g. the metric/km figures in “(1.607421875 19 km)”).

    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
     llama
    -0.07
    .utility
    -0.06
     Homes
    -0.06
     Tài
    -0.06
    icaret
    -0.06
     airlines
    -0.06
     LogManager
    -0.06
    ежду
    -0.06
    Č
    -0.06
    Forms
    -0.06
    POSITIVE LOGITS
    (simp
    0.06
    _BY
    0.06
    ="<<
    0.06
    	kfree
    0.06
    istrov
    0.06
    ."</
    0.06
    ất
    0.06
     nett
    0.06
    _X
    0.06
     neutral
    0.05
    Activations Density 0.004%

    No Known Activations