© 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. 8417
    Prev
    Next
    INDEX
    Explanations

    nave, ave, eve, ive, dove

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

    The neuron strongly activates on isolated capitalized tokens—proper names or uppercase acronyms.

    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
    อ
    -1.10
    la
    -0.95
    ss
    -0.94
    νης
    -0.94
    อบ
    -0.92
    nya
    -0.89
    อย่าง
    -0.87
    ness
    -0.82
    len
    -0.81
    อาจ
    -0.81
    POSITIVE LOGITS
    eh
    0.89
    hf
    0.89
    ">>
    0.84
    kym
    0.82
    ügen
    0.81
     Meksiku
    0.80
     Hervé
    0.78
    though
    0.77
    bria
    0.77
     témoignages
    0.77
    Activations Density 0.041%

    No Known Activations