© 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. 10-GEMMASCOPE-RES-131K
    7. 102244
    Prev
    Next
    INDEX
    Explanations

    which, whom

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

    This neuron is strongly activated by occurrences of the phrase “with which,” i.e. by that specific prepositional + relative-clause pattern.

    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
    狒
    -1.70
     nyní
    -1.66
     reminding
    -1.59
     niektórych
    -1.57
    てて
    -1.55
     kullanım
    -1.55
     belangrijke
    -1.52
     vormen
    -1.51
     gewoon
    -1.51
    Elektron
    -1.50
    POSITIVE LOGITS
     Moreover
    2.27
     It
    2.17
     Again
    2.14
     Therefore
    2.14
     Furthermore
    2.09
    ing
    1.98
    他
    1.91
     Hence
    1.88
     From
    1.86
     Thus
    1.86
    Activations Density 0.012%

    No Known Activations