© 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. 34-GEMMASCOPE-RES-131K
    7. 109474
    Prev
    Next
    INDEX
    Explanations

    is/are/was followed by description

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

    This neuron specifically detects numeric tokens and standalone numbers (e.g. measurements, counts, IDs) 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_34/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
     and
    -1.45
    (/^
    -1.41
    ,$_
    -1.22
    ?=
    -1.21
     it
    -1.20
    ciled
    -1.19
    imetry
    -1.18
    iamo
    -1.17
     karena
    -1.16
    agory
    -1.16
    POSITIVE LOGITS
     not
    1.59
     also
    1.26
     actually
    1.21
     in
    1.20
     located
    1.17
     positioned
    1.14
     labeled
    1.09
     devoid
    1.08
    我们要
    1.07
     made
    1.05
    Activations Density 0.107%

    No Known Activations