© 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. 10-GEMMASCOPE-RES-131K
    7. 130483
    Prev
    Next
    INDEX
    Explanations

    specific word beginnings

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

    The neuron fires on programming‐language tokens—especially code keywords and identifiers like “same” (padding) or “member” (in “member function”)—i.e. it detects code-specific terms in technical snippets.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     tries
    -1.84
     所以
    -1.75
     makes
    -1.73
     enormous
    -1.62
    z
    -1.59
     में
    -1.57
    ."
    -1.55
     मैं
    -1.51
     That
    -1.49
    
    -1.48
    POSITIVE LOGITS
    1
    2.30
    7
    1.96
    4
    1.94
    9
    1.93
    いえば
    1.73
    或者
    1.72
    呖
    1.66
    韪
    1.66
     paſſ
    1.65
    雠
    1.65
    Activations Density 0.000%

    No Known Activations