© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 39005
    Prev
    Next
    INDEX
    Explanations

    IDs, SHA, NAME, URL, Field

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

    The neuron responds to standalone numeric tokens (multi‐digit numbers), e.g. version numbers or counts.

    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
    Ook
    -1.20
    でもある
    -1.19
    修炼
    -1.18
    ༷
    -1.16
     vao
    -1.13
    DllImport
    -1.13
    thag
    -1.13
     ſet
    -1.12
     vora
    -1.11
     yyj
    -1.10
    POSITIVE LOGITS
     while
    1.30
     when
    1.20
    抑郁
    1.20
     now
    1.20
     helps
    1.13
     longtime
    1.13
     In
    1.05
     this
    1.03
     February
    1.03
     during
    1.02
    Activations Density 0.019%

    No Known Activations