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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 107019
    Prev
    Next
    INDEX
    Explanations

    salmon and trout

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

    fish-related words, especially trout and salmon.

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

    The main thing this neuron does is detect mentions of specific fish species (e.g., “salmon,” “trout”).

    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
    ntesis
    -0.73
     swamps
    -0.73
     Helden
    -0.73
     shipyard
    -0.73
     sunflowers
    -0.72
     végétale
    -0.71
     legis
    -0.71
     marshes
    -0.70
    Sunflower
    -0.69
    MPI
    -0.68
    POSITIVE LOGITS
     salmon
    2.73
     trout
    2.55
     Salmon
    2.34
    Salmon
    2.20
    salmon
    2.09
     Trout
    1.97
    trout
    1.79
    Trout
    1.72
     stream
    1.47
    Salmo
    1.43
    Activations Density 0.035%

    No Known Activations