© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 38387
    Prev
    Next
    INDEX
    Explanations

    version followed by equals or colon

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

    The neuron is picking out “Version” labels and their associated version‐number tokens in configuration or metadata text.

    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
     for
    -1.65
     all
    -1.46
     also
    -1.44
     many
    -1.41
     what
    -1.39
     if
    -1.34
     these
    -1.32
     several
    -1.29
     because
    -1.28
     about
    -1.27
    POSITIVE LOGITS
     اطلاعات
    1.13
     millennia
    1.09
    
    1.07
     bied
    1.05
     aktualisiert
    1.04
     vanguardia
    1.02
     interessante
    1.01
    Mitarbeiter
    1.01
     현재
    1.00
     борьбе
    1.00
    Activations Density 0.047%

    No Known Activations