© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 41375
    Prev
    Next
    INDEX
    Explanations

    whether to choose

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

    The neuron fires strongly on the auxiliary “will” (and related copula constructions like “is… a good choice for you”), i.e. on future‐tense or recommendation/prediction phrasing.

    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
     because
    -1.38
    oció
    -1.29
    ivamente
    -1.24
     lidé
    -1.22
    因為
    -1.20
     alip
    -1.20
    opie
    -1.19
     stihl
    -1.19
     satria
    -1.18
     erik
    -1.17
    POSITIVE LOGITS
     or
    1.52
     there
    1.45
     we
    1.20
    example
    1.19
    这样的
    1.18
    ↵
    1.16
     then
    1.13
     нач
    1.09
     with
    1.09
     тему
    1.08
    Activations Density 0.118%

    No Known Activations