© 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. 36105
    Prev
    Next
    INDEX
    Explanations

    collections of things

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

    The neuron is keyed to personal and conversational markers—especially first- and second-person pronouns (I, we, you), deictic words (this, your), and related function words that signal a personal or informal tone.

    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
     two
    -1.62
     three
    -1.41
     changes
    -1.33
     processes
    -1.27
     first
    -1.27
     when
    -1.22
     conditions
    -1.19
     procedures
    -1.19
     one
    -1.17
     on
    -1.16
    POSITIVE LOGITS
    这些
    1.53
     этих
    1.48
     которые
    1.39
    這些
    1.34
     них
    1.32
     onları
    1.31
     őket
    1.28
    攥
    1.27
     them
    1.23
     essas
    1.16
    Activations Density 0.113%

    No Known Activations