© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 72178
    Prev
    Next
    INDEX
    Explanations

    topics and discussions

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

    This neuron activates on domain-specific or technical topic words (often capitalized or in titles), such as specialized nouns naming subjects or headings.

    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
    indicates
    -1.09
    tells
    -1.03
     enthousi
    -1.01
    describing
    -1.01
     exagger
    -1.00
     appris
    -0.98
    大丈夫です
    -0.98
     ister
    -0.98
     procla
    -0.98
    -0.97
    POSITIVE LOGITS
     topic
    1.76
     discussed
    1.47
     discussion
    1.40
     discuss
    1.23
     topics
    1.16
    topic
    1.12
     discusses
    1.06
     analysis
    1.05
     subsection
    1.05
     article
    1.04
    Activations Density 0.077%

    No Known Activations