© 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. 71444
    Prev
    Next
    INDEX
    Explanations

    to + verb infinitive

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

    The neuron fires on words that express or govern ability/permission (e.g. “to,” “able,” “allow,” “make,” “be,” “can,” “trying”).

    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
     zü
    -1.02
     astonished
    -1.02
    cetak
    -1.02
     unieke
    -1.02
     grijs
    -0.98
     répondit
    -0.97
    iration
    -0.96
    atoms
    -0.96
     houten
    -0.96
     surprise
    -0.95
    POSITIVE LOGITS
    這麼
    1.54
    Such
    1.46
    这么
    1.38
     takto
    1.36
     such
    1.35
     these
    1.34
     столь
    1.34
    如此
    1.33
    這樣
    1.33
    これだけ
    1.27
    Activations Density 0.025%

    No Known Activations