© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 66189
    Prev
    Next
    INDEX
    Explanations

    early conversion

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

    The neuron detects MSBuild‐style property references, i.e. the “$(… )” variable interpolation syntax.

    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
     their
    -1.84
     at
    -1.42
     even
    -1.30
     both
    -1.28
     only
    -1.24
     not
    -1.21
     they
    -1.18
     membag
    -1.09
     it
    -1.08
     for
    -1.05
    POSITIVE LOGITS
    辊
    1.01
    𝟯
    1.01
    Descriere
    0.98
     crescita
    0.96
     alluminio
    0.94
     Arka
    0.92
     fijar
    0.92
     trattano
    0.92
    パンツ
    0.91
    AccessorType
    0.90
    Activations Density 0.000%

    No Known Activations