© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 58399
    Prev
    Next
    INDEX
    Explanations

    necessary and appropriate

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

    The neuron fires on longer, formally derived words—especially multi‐syllabic adjectives and adverbs (e.g. words ending in –ly, –ed, –ive, 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_34/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
    edor
    -0.83
    Incorrect
    -0.81
    优先
    -0.79
    iguar
    -0.79
     Incorrect
    -0.78
     Punto
    -0.78
    τι
    -0.76
     mudanças
    -0.75
     nedeniyle
    -0.75
     现代
    -0.75
    POSITIVE LOGITS
     narrowly
    1.42
     necessity
    1.32
     legitimate
    1.21
     proportionate
    1.19
     justified
    1.08
     necessary
    1.06
     proportional
    1.05
     narrow
    1.02
     reasonable
    1.00
     proportionality
    0.98
    Activations Density 0.015%

    No Known Activations