© 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. 76555
    Prev
    Next
    INDEX
    Explanations

    diatom shells

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

    scientific names of algae, diatoms, and other aquatic organisms.

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

    The neuron fires on specialized scientific or biological terminology (e.g. taxonomic names and technical vocabulary) commonly found in research‐article text.

    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
    bacterial
    -0.73
     Bacterial
    -0.72
     editor
    -0.70
     lobby
    -0.69
     microbial
    -0.69
     хо
    -0.68
    losse
    -0.68
    ybės
    -0.68
     bulletin
    -0.67
    роид
    -0.67
    POSITIVE LOGITS
     valve
    0.87
    raphe
    0.80
     frust
    0.80
     Valves
    0.77
     Frust
    0.76
    TEMA
    0.75
     VALVE
    0.74
     silic
    0.73
    centerY
    0.71
     作り方
    0.70
    Activations Density 0.026%

    No Known Activations