© 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. 70961
    Prev
    Next
    INDEX
    Explanations

    punctuation, initialisms, or days followed by context

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    の地
    -1.40
     this
    -1.31
    の方に
    -1.25
    cobacterium
    -1.22
    corrhi
    -1.20
    Almost
    -1.18
     lemah
    -1.17
    おわりに
    -1.16
    感じで
    -1.16
    Nearly
    -1.16
    POSITIVE LOGITS
    などが
    1.16
    <bos>
    1.04
    彻底
    1.02
    SenderId
    0.98
     hoi
    0.96
    がある
    0.96
    便捷
    0.96
    があり
    0.96
    があります
    0.95
    轍
    0.95
    Activations Density 0.000%

    No Known Activations