© 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. 28425
    Prev
    Next
    INDEX
    Explanations

    or, version, or

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

    The neuron activates on technical identifiers—short uppercase or mixed alphanumeric tokens and numbers—such as model numbers, version numbers, dates, and document or form codes.

    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
     você
    -1.05
    dfunding
    -1.04
    zeczytaj
    -1.03
     …
    -1.02
    |$.
    -1.01
     again
    -1.00
     повтор
    -1.00
     MUCH
    -0.99
     ตั้ง
    -0.98
    …
    -0.98
    POSITIVE LOGITS
     or
    2.63
     или
    1.47
     either
    1.45
    หรือ
    1.42
     preferably
    1.41
     Preferably
    1.33
     hoặc
    1.32
    either
    1.30
    preferably
    1.25
    或者
    1.24
    Activations Density 0.033%

    No Known Activations