© 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. 86952
    Prev
    Next
    INDEX
    Explanations

    set up and configured

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

    The neuron detects verbs presented in their past‐tense or past‐participle form, often indicating completed actions.

    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
    をします
    -1.81
     pracuje
    -1.65
    一定会
    -1.55
    〼
    -1.55
     cercana
    -1.54
     biela
    -1.53
     ciertos
    -1.53
     живут
    -1.51
     ")[
    -1.50
     stali
    -1.49
    POSITIVE LOGITS
     and
    2.13
     for
    1.87
    in
    1.54
     out
    1.48
     with
    1.48
     as
    1.44
     t
    1.36
     –
    1.35
    le
    1.34
    })$.
    1.34
    Activations Density 0.048%

    No Known Activations