© 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. 10-GEMMASCOPE-RES-131K
    7. 40646
    Prev
    Next
    INDEX
    Explanations

    abstract nouns and concepts

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

    This neuron primarily activates on numeric tokens or explicit number-like expressions 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_10/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
    抿
    -1.55
     on
    -1.52
     personnels
    -1.43
     frequentemente
    -1.41
    耵
    -1.40
    ↵
    -1.40
     normalmente
    -1.37
     ation
    -1.30
     I
    -1.30
     instead
    -1.30
    POSITIVE LOGITS
    although
    1.84
    This
    1.81
    These
    1.78
    ほとんど
    1.57
    Although
    1.57
    Since
    1.55
     these
    1.54
    urbing
    1.52
    Many
    1.52
    Despite
    1.50
    Activations Density 0.085%

    No Known Activations