© 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. 66331
    Prev
    Next
    INDEX
    Explanations

    abstract properties and states

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

    The neuron activates on numeric tokens, identifying numbers (e.g. section labels, measurements, reference numbers) 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
    StoryboardSegue
    -1.27
     pentru
    -1.24
    了一会儿
    -1.20
    delse
    -1.18
    niej
    -1.17
     atau
    -1.16
     だっ
    -1.14
    部は
    -1.13
    了下去
    -1.12
    corrhi
    -1.12
    POSITIVE LOGITS
     fact
    1.91
     ability
    1.70
     inability
    1.49
     to
    1.36
     presence
    1.36
     interplay
    1.31
     Tatsache
    1.26
     availability
    1.24
     Fähigkeit
    1.21
     Grüßen
    1.19
    Activations Density 0.089%

    No Known Activations