© 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. 25448
    Prev
    Next
    INDEX
    Explanations

    creation and development processes

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

    The neuron strongly responds to verbs—especially present‐tense and gerund forms—that describe actions or processes in technical (often code or documentation) contexts.

    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
    個人的には
    -1.26
     honte
    -1.22
     masculinos
    -1.16
    などが
    -1.12
    равда
    -1.11
     wartości
    -1.08
     deportivos
    -1.05
    それとも
    -1.05
     całej
    -1.04
    まさか
    -1.04
    POSITIVE LOGITS
     and
    1.41
     This
    1.33
    apiVersion
    1.00
     Ultimately
    0.94
     générations
    0.93
    Transcrip
    0.90
     Actually
    0.90
     Subsequently
    0.89
    ânica
    0.88
    meteries
    0.87
    Activations Density 0.068%

    No Known Activations