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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 31-GEMMASCOPE-2-RES-262K
    7. 13935
    Prev
    Next
    INDEX
    Explanations

    evaluative adjectives or descriptors

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

    the neuron detects technical trade-off language—phrases describing performance/resource compromises (e.g., reducing memory at the cost of slightly degraded model quality).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Existem
    0.42
     Ð
    0.40
     ď
    0.39
     포함
    0.39
     Þ
    0.39
     в
    0.38
     Über
    0.38
     particulier
    0.38
     À
    0.38
     in
    0.38
    POSITIVE LOGITS
     certes
    0.40
     tillage
    0.36
     expts
    0.36
    FAPI
    0.32
    RefreshToken
    0.32
    ρεία
    0.32
    Beverungen
    0.32
     calyc
    0.31
     sicherlich
    0.31
     уены
    0.31
    Activations Density 0.327%

    No Known Activations