© 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. 68496
    Prev
    Next
    INDEX
    Explanations

    `post`, `In`, `Analysis`, `Using`

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

    The neuron is tuned to detect technical measurement expressions—that is, numeric values, fractions, units, and other dimension‐specifying tokens.

    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
     бампер
    -1.04
    czną
    -1.03
    yoruz
    -1.03
    ificity
    -1.02
     özellik
    -1.02
    tzmann
    -1.01
    sci
    -1.00
    CARACTER
    -0.99
    Source
    -0.96
    Zj
    -0.96
    POSITIVE LOGITS
     this
    1.15
    ÜN
    1.05
     succede
    1.01
     médié
    1.01
    pendicular
    1.01
    chdir
    1.00
     گیرد
    0.99
    ׇ
    0.98
    如图
    0.98
    Zitat
    0.98
    Activations Density 0.057%

    No Known Activations