© 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. 14609
    Prev
    Next
    INDEX
    Explanations

    i

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

    The neuron activates on standalone numeric tokens (i.e. sequences of digits).

    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
     on
    -1.41
     with
    -1.27
     an
    -1.21
     both
    -1.13
     assessing
    -1.12
    vably
    -1.12
     keperluan
    -1.12
     about
    -1.10
    ieuses
    -1.10
     observa
    -1.09
    POSITIVE LOGITS
     gäng
    1.26
    することができます
    1.26
     it
    1.22
     konkrete
    1.16
     Unterscheidung
    1.16
    部分を
    1.14
     resolução
    1.13
     différents
    1.13
     différentes
    1.12
    つも
    1.11
    Activations Density 0.231%

    No Known Activations