© 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. 48840
    Prev
    Next
    INDEX
    Explanations

    ends in io

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

    The neuron fires on any token containing the letter sequence “io,” effectively detecting that substring wherever it appears.

    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
     varied
    -1.00
     oddly
    -0.94
     something
    -0.94
     exact
    -0.92
     great
    -0.91
     surprisingly
    -0.89
     valued
    -0.85
     weirdly
    -0.84
    解釈
    -0.83
     identify
    -0.82
    POSITIVE LOGITS
     ness
    0.94
     intermin
    0.91
    cination
    0.87
    宙
    0.82
     Smoothie
    0.79
    Assist
    0.79
    beiro
    0.79
    NEXT
    0.79
    荒
    0.79
    дио
    0.78
    Activations Density 0.074%

    No Known Activations