© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 58676
    Prev
    Next
    INDEX
    Explanations

    bit size

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

    This neuron activates strongly on numeric tokens (e.g., integer literals, multi-digit numbers, bit-widths, years) wherever they appear in code or text.

    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
     never
    -2.34
    never
    -1.95
    Never
    -1.84
     Never
    -1.69
     никогда
    -1.59
     nigdy
    -1.51
     nunca
    -1.36
     nooit
    -1.30
     nikdy
    -1.27
     jamais
    -1.27
    POSITIVE LOGITS
     bit
    2.86
    bit
    2.81
    Bit
    2.59
     bits
    2.58
    bits
    2.58
     Bits
    2.41
     Bit
    2.38
    Bits
    2.30
    BIT
    2.09
    BITS
    1.92
    Activations Density 0.074%

    No Known Activations