© 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. 22-GEMMASCOPE-RES-131K
    7. 138
    Prev
    Next
    INDEX
    Explanations

    auto parts and replacements

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

    The neuron activates on structured product-specification labels and table headings (e.g. “Compatibility,” “Year,” “Model,” “Engine,” etc.).

    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
    corbic
    -1.52
     transportar
    -1.38
     ventajas
    -1.25
    ₆
    -1.23
    ERÍA
    -1.18
     céle
    -1.17
    ynchronously
    -1.16
    řejmě
    -1.16
    InkWell
    -1.16
    larının
    -1.15
    POSITIVE LOGITS
     replacement
    1.98
     replacements
    1.74
     replacing
    1.46
     replace
    1.43
    replacement
    1.41
    your
    1.39
     your
    1.37
     sostitu
    1.32
     originali
    1.30
     tested
    1.27
    Activations Density 0.039%

    No Known Activations