© 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. 38237
    Prev
    Next
    INDEX
    Explanations

    Slavic names/locations

    np_max-act · gemini-2.0-flash

    The neuron activates on Russian place‐name words using adjectival suffixes (e.g. –insky, –ovsky, –skaya) that mark administrative districts and localities.

    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
    摸
    -0.08
    '''↵↵
    -0.07
     меж
    -0.06
    üyorum
    -0.06
     QName
    -0.06
    Pixels
    -0.06
    (SIG
    -0.06
     Kon
    -0.06
     '=
    -0.06
    measurement
    -0.06
    POSITIVE LOGITS
     Tracker
    0.07
    ways
    0.07
    	retval
    0.07
    ric
    0.07
    cular
    0.06
     良
    0.06
    folk
    0.06
    ingly
    0.06
    ovsky
    0.06
    /rfc
    0.06
    Activations Density 0.057%

    No Known Activations