© 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. 10-GEMMASCOPE-RES-131K
    7. 100670
    Prev
    Next
    INDEX
    Explanations

    connecting phrases

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

    The neuron strongly activates on occurrences of the word “with.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    铪
    -1.92
    com
    -1.59
    鹩
    -1.59
    ;:
    -1.53
    犼
    -1.52
    dans
    -1.48
    ;-
    -1.48
    饪
    -1.48
     -.
    -1.45
     pretože
    -1.45
    POSITIVE LOGITS
     όμως
    1.80
    amanan
    1.60
     troubled
    1.60
    </h2>
    1.59
    Each
    1.57
     painfully
    1.54
    後來
    1.45
    énagement
    1.45
    Then
    1.44
    Every
    1.44
    Activations Density 0.100%

    No Known Activations