© 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. 34-GEMMASCOPE-RES-131K
    7. 118195
    Prev
    Next
    INDEX
    Explanations

    numbers and closing punctuation

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

    The neuron activates on standalone numeric tokens (track numbers, time stamps, durations, and other digits).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    multi-digit numerals (especially three- or four-digit numbers, such as times, IDs, or other numeric metadata).

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    numeric values that appear after time durations or system call constants, particularly in structured lists or code.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    :
    -1.20
     from
    -1.17
     that
    -1.09
    ,
    -1.08
    詫
    -1.06
     will
    -1.02
    IConfiguration
    -1.02
     of
    -1.00
     or
    -0.98
     as
    -0.98
    POSITIVE LOGITS
    ↵
    1.23
    Noter
    1.16
     quantidade
    1.13
     ilości
    1.12
     relazione
    1.12
    Kleur
    1.08
     überprü
    1.07
    Materiale
    1.05
    lograph
    1.04
     estructuras
    1.03
    Activations Density 0.033%

    No Known Activations