© 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. Gemma-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 5619
    Prev
    Next
    INDEX
    Explanations

    words that appear to be location names or government positions within a city or state

    oai_token-act-pair · gemini-2.0-flash

    city government/politics

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     municipal
    -1.10
    municipal
    -1.02
     municipality
    -0.97
    Municipal
    -0.95
     MUNICIPAL
    -0.91
     municip
    -0.89
     Municipal
    -0.89
     Municipality
    -0.86
     municipales
    -0.83
     council
    -0.82
    POSITIVE LOGITS
    :+:
    0.62
    IUrlHelper
    0.54
    पया
    0.54
    Hochspringen
    0.52
     stanu
    0.52
    ativement
    0.51
    UnusedPrivate
    0.51
    webElement
    0.50
     Gobierno
    0.49
    JspWriter
    0.48
    Activations Density 0.918%

    No Known Activations