© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 93389
    Prev
    Next
    INDEX
    Explanations

    tuple or list separators

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

    The neuron fires on numeric literals (tokens consisting of digits or 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_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
     is
    -1.48
    那我
    -1.30
    other
    -1.27
    1
    -1.27
    then
    -1.23
     dalších
    -1.23
     getrock
    -1.21
     ]
    -1.19
     was
    -1.18
    tion
    -1.16
    POSITIVE LOGITS
     maakte
    1.32
    彣
    1.29
    tyfik
    1.28
    ldern
    1.27
     schreef
    1.27
     medalla
    1.26
    cyj
    1.23
     guber
    1.22
     plena
    1.21
     mecan
    1.20
    Activations Density 0.011%

    No Known Activations