© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 100601
    Prev
    Next
    INDEX
    Explanations

    to + verb infinitive

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

    The neuron is primarily detecting the infinitive marker “to” (including its use in prepositional and infinitive constructions).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    man
    -1.96
    \
    -1.74
    珦
    -1.63
     that
    -1.61
     after
    -1.53
    ",
    -1.51
    zieć
    -1.50
     skolen
    -1.48
    k
    -1.46
    vereignty
    -1.45
    POSITIVE LOGITS
    椟
    1.62
     somewhat
    1.60
    ѻ
    1.53
    
    1.52
    giu
    1.51
    vonne
    1.48
    はかなり
    1.48
     nommé
    1.42
     largely
    1.41
    ⓵
    1.40
    Activations Density 0.005%

    No Known Activations