© 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. 22-GEMMASCOPE-RES-131K
    4. 71553
    Prev
    Next
    INDEX
    Explanations

    new customers, copying DNA, Canyon State

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

    technical or professional terms, particularly in scientific, medical, and technological contexts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    This neuron primarily activates on longer, multi-syllable content words (especially technical or formal nouns and verbs).

    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
     dépenses
    -1.21
    扨
    -1.16
     WHILE
    -1.14
     joyous
    -1.14
     Infór
    -1.11
     jezt
    -1.09
    鶫
    -1.08
     MICH
    -1.07
     téměř
    -1.07
     localización
    -1.06
    POSITIVE LOGITS
     just
    1.32
     For
    1.27
     Unlike
    1.26
     When
    1.26
     From
    1.20
     With
    1.19
     so
    1.17
     from
    1.16
     their
    1.15
     only
    1.15
    Activations Density 0.103%

    No Known Activations