© 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. 22-GEMMASCOPE-RES-131K
    4. 67570
    Prev
    Next
    INDEX
    Explanations

    corresponding or matching

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

    The neuron lights up on dense, domain-specific technical jargon—especially long, compound terms and specialized nouns.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     towards
    -0.83
    尤其是
    -0.81
    Towards
    -0.72
    經典
    -0.71
    特别是
    -0.70
    歡
    -0.70
    をお願い
    -0.70
     headed
    -0.70
     médic
    -0.69
    Muito
    -0.69
    POSITIVE LOGITS
     corresponding
    1.62
    corresponding
    1.41
     matching
    1.41
     mirroring
    1.38
    matching
    1.34
    same
    1.30
     same
    1.28
     counterpart
    1.21
     dezelfde
    1.20
     paralel
    1.18
    Activations Density 0.147%

    No Known Activations