© 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. 33632
    Prev
    Next
    INDEX
    Explanations

    soft descriptions and materials

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

    The neuron activates specifically on the adjective “soft.”

    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
     
    -2.72
    ない
    -2.36
    ↵
    -2.23
    腚
    -2.23
    这般
    -2.20
    呿
    -2.13
     governador
    -2.09
    WAUKEE
    -2.00
    皷
    -1.98
    呴
    -1.95
    POSITIVE LOGITS
    7
    2.91
    4
    2.84
    6
    2.80
    9
    2.72
    0
    2.67
    5
    2.63
    How
    2.58
    We
    2.58
    One
    2.55
    When
    2.55
    Activations Density 0.007%

    No Known Activations