© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 16242
    Prev
    Next
    INDEX
    Explanations

    linking verbs and auxiliary verbs

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

    The neuron strongly activates on numeric and measurement‐style tokens (e.g. dosages, counts, statistical values) appearing in technical or scientific contexts.

    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
     OGSÅ
    -1.47
     verhind
    -1.35
    niego
    -1.24
     PROCEEDINGS
    -1.20
    Oltre
    -1.20
    ᠌
    -1.19
     sánchez
    -1.17
     récol
    -1.16
     and
    -1.16
    internalType
    -1.16
    POSITIVE LOGITS
     as
    1.64
     on
    1.38
     at
    1.36
     consequently
    1.27
     it
    1.23
    代わりに
    1.11
     there
    1.07
    同様に
    1.05
     its
    1.05
     old
    1.03
    Activations Density 0.044%

    No Known Activations