© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 75319
    Prev
    Next
    INDEX
    Explanations

    increased heart rate, hair thickness, height, congenital HL

    np_acts-logits-general · gemini-2.5-flash-lite

    statements or descriptions where one subject has or possesses certain characteristics or properties.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron selectively fires on numeric measurement tokens (especially numbers, including decimals) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     which
    -1.41
    zdo
    -1.38
    mẫu
    -1.30
     οποίο
    -1.26
     usługi
    -1.23
     wensen
    -1.23
    𖥦
    -1.22
    ugía
    -1.20
    corrhi
    -1.20
    Febru
    -1.20
    POSITIVE LOGITS
     is
    1.70
     again
    1.62
     similar
    1.31
    Again
    1.27
    has
    1.19
    3
    1.16
    again
    1.16
    is
    1.16
     Again
    1.13
    t
    1.11
    Activations Density 0.016%

    No Known Activations