© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 65923
    Prev
    Next
    INDEX
    Explanations

    intramolecular, classical, glyco-

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

    scientific or technical terms in academic or technical documents.

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

    The neuron is picking out morpheme fragments (prefixes or suffixes) of technical or scientific multi-part terms (e.g. chemical names, process names, or specialized terminology).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     Màu
    -0.92
     Quý
    -0.88
    ėmis
    -0.88
    Sino
    -0.85
    PROBLE
    -0.85
     mathematicians
    -0.84
     espirituales
    -0.84
     gillar
    -0.83
     ikiwa
    -0.82
    結果は
    -0.78
    POSITIVE LOGITS
     < 
    1.05
     SIX
    0.96
    NH
    0.92
    Scheme
    0.91
    PTION
    0.91
     prega
    0.90
    ränke
    0.89
     muka
    0.88
    佝
    0.87
    viated
    0.86
    Activations Density 0.000%

    No Known Activations