© 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. 9529
    Prev
    Next
    INDEX
    Explanations

    finding following investigations

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

    The neuron is specialized to pick out numeric tokens—counts, ages, times, measurements—wherever digits appear 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
     Ejemplo
    -1.39
     positifs
    -1.31
     congrès
    -1.26
     véritables
    -1.25
    ovým
    -1.22
     ymm
    -1.21
     modèles
    -1.20
     efficacité
    -1.18
     prachtige
    -1.18
    ojo
    -1.18
    POSITIVE LOGITS
     it
    1.55
     no
    1.53
     community
    1.31
     said
    1.31
     he
    1.27
     police
    1.27
     follow
    1.26
     avoid
    1.20
     check
    1.20
     two
    1.20
    Activations Density 0.016%

    No Known Activations