© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 45668
    Prev
    Next
    INDEX
    Explanations

    mel

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

    The neuron fires strongly on any token containing the substring “Mel,” i.e. words beginning with or including “Mel.”

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    ↵↵
    -3.30
    翆
    -2.80
     anderen
    -2.80
    
    -2.78
    
    -2.72
     バッジ
    -2.69
    􀂃
    -2.67
    灩
    -2.66
    曖昧さ
    -2.53
     mesmerizing
    -2.52
    POSITIVE LOGITS
    1
    2.75
    相続
    2.48
    鸶
    2.39
    颙
    2.33
    我會
    2.31
    我没
    2.30
     apparaître
    2.27
    ストリー
    2.20
    我这
    2.20
    老闆
    2.20
    Activations Density 0.004%

    No Known Activations