© 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. 95484
    Prev
    Next
    INDEX
    Explanations

    chart positions and awards

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

    The neuron fires strongly on the conventional past‐tense “be” constructions (especially “was”) that introduce release or chart‐performance statements (e.g. “It was released…,” “The first single…was…,” “However…”).

    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
    advantage
    -1.26
     their
    -1.23
     new
    -1.22
    íba
    -1.10
    )।
    -1.10
    aS
    -1.08
     know
    -1.06
    ünstig
    -1.02
    tadores
    -0.99
    nejšie
    -0.99
    POSITIVE LOGITS
     sold
    1.30
     was
    1.30
     popularity
    1.27
     success
    1.17
     муки
    1.16
     oração
    1.11
    логии
    1.10
     insured
    1.09
     reson
    1.08
     successful
    1.05
    Activations Density 0.070%

    No Known Activations