© 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. 70928
    Prev
    Next
    INDEX
    Explanations

    medical conditions and specific objects

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

    scientific terms and biological/medical terminology, especially when related to specialized cells, proteins, or materials.

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

    The neuron fires on long, multi‐syllabic technical or scientific jargon terms.

    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
     pán
    -1.43
     kabát
    -1.41
     rám
    -1.25
     kré
    -1.21
     combattre
    -1.19
     luxus
    -1.18
     maíz
    -1.18
     flesta
    -1.14
     tazas
    -1.11
     pól
    -1.10
    POSITIVE LOGITS
    Informazioni
    1.05
     or
    0.97
     спросил
    0.96
    Escrito
    0.96
    ssohn
    0.96
    いくつかの
    0.96
     şeyi
    0.95
    Presentación
    0.95
    Вопрос
    0.95
    tgt
    0.94
    Activations Density 0.192%

    No Known Activations