© 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. 34-GEMMASCOPE-RES-131K
    7. 99211
    Prev
    Next
    INDEX
    Explanations

    that or think

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

    The neuron activates on occurrences of the subordinating conjunction “that,” especially when it follows verbs or nouns of belief, assumption, suggestion, 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
     zarówno
    -1.13
     både
    -1.13
    ります
    -1.09
    Istorija
    -1.09
     چه
    -1.08
    même
    -1.08
     навіть
    -1.07
    ரும்
    -1.04
    reo
    -1.02
    imensional
    -1.02
    POSITIVE LOGITS
     because
    2.36
     only
    2.30
     all
    2.09
     since
    1.83
     there
    1.77
     it
    1.64
    Only
    1.63
     everything
    1.59
     you
    1.58
    Since
    1.58
    Activations Density 0.065%

    No Known Activations