© 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. 103346
    Prev
    Next
    INDEX
    Explanations

    virtual, animal, cow

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

    This neuron activates on tokens involved in code-level assignments or definitions—i.e. variable or parameter names (and related keywords) at assignment sites.

    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
     both
    -0.84
     Current
    -0.79
     Mr
    -0.79
    琐
    -0.79
     тепла
    -0.78
     these
    -0.78
    both
    -0.77
     Other
    -0.77
    either
    -0.77
     They
    -0.76
    POSITIVE LOGITS
    HANG
    0.95
     RoHS
    0.93
    显然
    0.91
    謌
    0.91
     EXCELL
    0.89
    TNF
    0.88
    Ellen
    0.85
    SafeArea
    0.85
    uttgart
    0.85
    ilita
    0.85
    Activations Density 0.179%

    No Known Activations