© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 104890
    Prev
    Next
    INDEX
    Explanations

    comes to

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

    The neuron specifically detects the phrase “comes to” (i.e. the word comes followed by to).

    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
    </i>
    -2.11
    ");
    
    -1.98
     všechny
    -1.97
    ంలో
    -1.92
    』。
    -1.88
    
    -1.88
    特長
    -1.86
    )$.
    -1.81
     músculos
    -1.74
    ↑
    -1.73
    POSITIVE LOGITS
     to
    3.19
    </strong>
    2.27
    ный
    1.96
    “
    1.91
    
    1.88
    PostInfinity
    1.83
    ”
    1.83
    il
    1.82
     Then
    1.80
     you
    1.77
    Activations Density 0.006%

    No Known Activations