© 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. 90408
    Prev
    Next
    INDEX
    Explanations

    heaven or heavens

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

    The neuron is highly selective for occurrences of the word “Heaven.”

    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
    
    -2.66
    </strong>
    -2.61
    “
    -2.58
     può
    -2.48
    ?"
    -2.44
     ';
    -2.27
    綃
    -2.17
    .”
    -2.17
    头部
    -2.16
    ?’
    -2.14
    POSITIVE LOGITS
    .
    2.80
    砹
    2.70
     otras
    2.59
     "
    2.56
    ization
    2.50
    ⟞
    2.50
    崀
    2.50
    </b>
    2.34
    <bos>
    2.33
    岜
    2.27
    Activations Density 0.001%

    No Known Activations