© 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. 40284
    Prev
    Next
    INDEX
    Explanations

    facilitating communication or providing features

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

    The neuron strongly activates on technical or domain-specific jargon words (e.g. specialized scientific or engineering terms).

    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
     presumably
    -1.31
     both
    -1.30
     should
    -1.26
     undoubtedly
    -1.23
     obviously
    -1.23
     or
    -1.21
     serves
    -1.20
    💕
    -1.20
    ８
    -1.20
     takes
    -1.18
    POSITIVE LOGITS
     créer
    1.48
    沢山
    1.38
     faciliter
    1.31
    細かく
    1.28
    毳
    1.25
     semelh
    1.24
     supprimer
    1.23
    慢慢的
    1.23
     alguns
    1.22
    柔ら
    1.22
    Activations Density 0.042%

    No Known Activations