© 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. 58078
    Prev
    Next
    INDEX
    Explanations

    mm

    np_max-act · gemini-2.0-flash

    The neuron activates on numeric measurements in millimeters (e.g., “1 mm,” “2.375 mm,” “4 mm”), so it’s detecting dimension values specified in mm.

    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
     Couch
    -0.08
     XCT
    -0.07
    COD
    -0.07
    (":/
    -0.07
    483
    -0.07
     çağ
    -0.06
    god
    -0.06
    兴
    -0.06
     dracon
    -0.06
    erç
    -0.06
    POSITIVE LOGITS
     mm
    0.12
    mm
    0.12
    MM
    0.10
    -mm
    0.10
    HH
    0.09
    (mm
    0.08
     Miller
    0.08
     MM
    0.08
     Hamm
    0.08
     AA
    0.07
    Activations Density 0.010%

    No Known Activations