© 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. 22-GEMMASCOPE-RES-131K
    7. 35338
    Prev
    Next
    INDEX
    Explanations

    contrasts, but

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

    The neuron responds strongly to common English function words—especially discourse markers (conjunctions, pronouns, and demonstratives) that structure or connect sentences.

    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
    зик
    -0.89
    èmes
    -0.85
    )=>{
    -0.81
    答え
    -0.79
    ++;
    -0.77
     Vorlage
    -0.77
     夢
    -0.75
    \}^{
    -0.75
     coment
    -0.74
     Vibes
    -0.73
    POSITIVE LOGITS
     comunque
    1.16
    Fortunately
    0.95
     anyway
    0.94
    それでも
    0.93
     testGet
    0.92
     Fortunately
    0.91
     OK
    0.91
     fortunately
    0.85
     glad
    0.83
     okay
    0.82
    Activations Density 0.052%

    No Known Activations