© 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. 34585
    Prev
    Next
    INDEX
    Explanations

    instructions and methods

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

    The neuron activates on infinitive‐style verb phrases in procedural or instructional contexts (e.g. “to raise,” “used to heat,” “core muscles”).

    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
    ございます
    -0.99
    旆
    -0.90
     Barca
    -0.90
    Wert
    -0.89
    Takk
    -0.88
     teater
    -0.88
    רקע
    -0.86
     Перейти
    -0.85
    шел
    -0.84
    Geor
    -0.84
    POSITIVE LOGITS
     if
    1.48
     this
    1.32
     even
    1.19
     such
    1.18
     dabei
    1.16
     to
    1.10
     so
    1.06
    Dabei
    1.03
     doing
    1.02
     melakukannya
    1.02
    Activations Density 0.066%

    No Known Activations