© 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. 117333
    Prev
    Next
    INDEX
    Explanations

    chemical or biological strains

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

    legal document citations, technical codes, and programming directives.

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

    The neuron fires on standalone numerals (e.g., version numbers, IDs, citation numbers).

    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
    tigkeits
    -1.02
    ahme
    -0.96
     its
    -0.93
    änder
    -0.92
    ΰ
    -0.92
     they
    -0.88
     who
    -0.87
    
    -0.86
    PROBE
    -0.85
     my
    -0.84
    POSITIVE LOGITS
     låter
    1.09
     acontec
    0.90
     bisa
    0.90
     Джо
    0.88
     aident
    0.88
     =
    0.82
    zünd
    0.82
    投手
    0.81
    IllegalArgument
    0.81
     by
    0.81
    Activations Density 0.000%

    No Known Activations