© 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. 34-GEMMASCOPE-RES-131K
    7. 87117
    Prev
    Next
    INDEX
    Explanations

    don't just

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

    The neuron strongly activates on adverbs and negation or emphasis words that qualify or limit a statement (e.g. don’t, never, only, just, really).

    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
     sudah
    -1.05
    libert
    -0.98
    PENUTUP
    -0.96
     nicht
    -0.96
     parece
    -0.94
     tidak
    -0.93
     quitté
    -0.92
     tampaknya
    -0.92
    ("")]
    -0.91
     không
    -0.86
    POSITIVE LOGITS
     seek
    1.37
     seeking
    1.24
     fear
    1.23
     seeks
    1.20
     ever
    1.16
     just
    1.10
     merely
    1.10
     Verbreitung
    1.07
     shy
    1.06
     ask
    1.05
    Activations Density 0.033%

    No Known Activations