© 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. 61357
    Prev
    Next
    INDEX
    Explanations

    down to

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

    The main thing this neuron does is identify numeric tokens (especially decimal measurements or quantities).

    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
     increase
    -1.84
    お話
    -1.55
    多かった
    -1.45
    歐洲
    -1.41
    ｐ
    -1.36
    ⟢
    -1.34
    hamilan
    -1.34
     promising
    -1.34
     up
    -1.34
    棪
    -1.33
    POSITIVE LOGITS
    ",
    1.85
    ),
    1.62
    },
    1.56
    \",
    1.51
     down
    1.51
    -
    1.49
     luxuri
    1.48
    颡
    1.39
    m
    1.37
    нія
    1.33
    Activations Density 0.016%

    No Known Activations