© 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. 104252
    Prev
    Next
    INDEX
    Explanations

    prefixes of names/codes

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

    The neuron consistently spikes on two‐letter nucleotide bigrams (dinucleotides) in genomic sequence listings.

    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
     in
    -1.86
     during
    -1.85
     as
    -1.71
     while
    -1.69
     for
    -1.67
     or
    -1.64
     would
    -1.64
     because
    -1.59
     when
    -1.58
     had
    -1.57
    POSITIVE LOGITS
     unbelievably
    1.52
     quintessential
    1.50
     amazingly
    1.48
     abd
    1.38
    撂
    1.32
     exceptionally
    1.31
     opulent
    1.31
     flamboyant
    1.30
     horrifying
    1.28
     alluring
    1.28
    Activations Density 0.000%

    No Known Activations