© 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. 34-GEMMASCOPE-RES-131K
    7. 91586
    Prev
    Next
    INDEX
    Explanations

    generation, creation, and processes

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

    The neuron fires on technical nouns naming processes or actions—especially words ending in “-ion” or “-ment” (e.g. manipulation, conversion, creation, analysis, generation).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     of
    -1.43
    _
    -1.23
    '
    -1.03
    ...
    -1.01
    els
    -0.99
     combating
    -0.96
     eve
    -0.96
    ous
    -0.94
     embraces
    -0.94
     gewä
    -0.94
    POSITIVE LOGITS
     and
    1.79
     schop
    1.20
     during
    1.11
     process
    1.08
     through
    1.07
    ability
    1.06
     및
    1.05
     in
    1.04
     methods
    1.03
     sarana
    1.03
    Activations Density 0.384%

    No Known Activations