© 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. 55822
    Prev
    Next
    INDEX
    Explanations

    Great, Related, Callum

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

    The neuron detects standalone numeric tokens.

    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
     (
    -1.26
     “
    -1.13
     FOR
    -1.00
     Surprisingly
    -0.97
    ~~~~~~~~~~~~~~~~
    -0.94
     because
    -0.94
     EVERY
    -0.94
     (“
    -0.90
     had
    -0.87
     gets
    -0.87
    POSITIVE LOGITS
     horrid
    1.16
     splendid
    1.16
     immense
    1.13
     marvelous
    1.12
     absurd
    1.12
     broader
    1.12
     monstrous
    1.11
     thậm
    1.09
     pivotal
    1.09
     dreadful
    1.07
    Activations Density 0.000%

    No Known Activations