© 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. 54435
    Prev
    Next
    INDEX
    Explanations

    specific nouns and descriptors

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

    The neuron spikes on content‐heavy, domain-specific nouns or technical keywords that carry the main topic of a passage.

    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
     with
    -1.24
     With
    -1.17
     while
    -1.13
     Unlike
    -1.03
     Before
    -1.02
     While
    -1.02
     first
    -1.01
     Also
    -1.00
     to
    -0.99
     three
    -0.99
    POSITIVE LOGITS
     niedersachsen
    1.15
    艄
    1.13
    沵
    1.12
     gewi
    1.09
     satria
    1.08
    premier
    1.07
     magnificent
    1.06
    𓆏
    1.05
    ſſen
    1.05
    鲼
    1.05
    Activations Density 0.034%

    No Known Activations