© 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. 22-GEMMASCOPE-RES-131K
    7. 67822
    Prev
    Next
    INDEX
    Explanations

    contrasts and specific states

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

    The neuron activates on comparative or contrastive statements—especially ones that weigh quantities or likelihoods against each other (e.g. “but,” “while,” “more,” “only,” numeric terms, etc.).

    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
     temi
    -0.82
     Город
    -0.79
    боль
    -0.79
    Maxim
    -0.78
    ktionen
    -0.75
     Extracts
    -0.75
    asmuch
    -0.74
    淵
    -0.73
     inclusion
    -0.73
    Teach
    -0.73
    POSITIVE LOGITS
     unlike
    1.21
     physical
    1.21
    Unlike
    1.17
     compared
    1.13
     can
    1.09
     infinitely
    1.06
    unlike
    0.95
     cost
    0.93
    不像
    0.93
     costs
    0.92
    Activations Density 0.069%

    No Known Activations