© 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. 34-GEMMASCOPE-RES-131K
    4. 115866
    Prev
    Next
    INDEX
    Explanations

    parts of text or code

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

    The neuron fires on numeric tokens—i.e. digits or numbers (often in code comments or literals), such as “1976,” “1664,” “1000,” etc.

    oai_token-act-pair · o4-miniTriggered 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
    ৳
    -0.95
     рубе
    -0.90
    читься
    -0.87
     rozwo
    -0.80
     женские
    -0.80
     Consumer
    -0.79
    caccio
    -0.79
     insufic
    -0.77
     keeping
    -0.77
    ﷼
    -0.77
    POSITIVE LOGITS
     LOOKS
    0.90
     gigabyte
    0.88
    lcccccc
    0.88
     SEVENT
    0.85
     TRUTH
    0.84
     certamente
    0.83
     의미
    0.83
     risposte
    0.81
     sfida
    0.81
    TER
    0.81
    Activations Density 0.008%

    No Known Activations