© 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. 92978
    Prev
    Next
    INDEX
    Explanations

    * and # characters

    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
     was
    -1.30
    ーズン
    -1.18
    atori
    -1.16
    ༵
    -1.16
     indicates
    -1.15
    jaqueta
    -1.13
     went
    -1.13
    ATION
    -1.12
    깐
    -1.12
     him
    -1.11
    POSITIVE LOGITS
    o
    1.41
     Hinblick
    1.33
    ll
    1.24
     peculiarly
    1.15
    d
    1.08
     отношение
    1.08
    ]
    1.03
    に関しては
    1.01
    针对
    1.00
     déplacer
    0.99
    Activations Density 0.003%

    No Known Activations