© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 124694
    Prev
    Next
    INDEX
    Explanations

    existing projects and assets

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

    The neuron fires on mentions of concrete entities or structures—generic nouns naming existing objects or resources (e.g. “table,” “key,” “processes,” “site,” “measures”) in technical or procedural contexts.

    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
     again
    -1.32
    sprogram
    -1.29
    なこと
    -1.19
     arbe
    -1.18
    Ikus
    -1.16
    AndPassword
    -1.16
     Kruse
    -1.16
     }_{
    -1.15
     PLATFORM
    -1.14
    薷
    -1.13
    POSITIVE LOGITS
    existing
    1.75
     existing
    1.69
     Existing
    1.63
    Existing
    1.44
     bestehenden
    1.38
    现有
    1.37
    伱
    1.25
    既存
    1.24
     existentes
    1.23
     existente
    1.23
    Activations Density 0.034%

    No Known Activations