© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 36055
    Prev
    Next
    INDEX
    Explanations

    dependence, fires, interest, cortisol

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

    The neuron responds to substantive, topic-bearing content words—domain-specific nouns and verbs that carry the core subject matter of a passage.

    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
     or
    -0.97
     for
    -0.91
     commensurate
    -0.87
     helps
    -0.87
    An
    -0.85
     розташо
    -0.85
     Electro
    -0.84
     relinqu
    -0.84
     она
    -0.83
    みたいに
    -0.83
    POSITIVE LOGITS
     already
    1.43
    があり
    1.38
    があって
    1.32
     was
    1.26
     apparently
    1.13
    已经
    1.11
    に対する
    1.06
    がありました
    1.05
    がありますが
    1.02
    keted
    1.02
    Activations Density 0.071%

    No Known Activations