© 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. 104662
    Prev
    Next
    INDEX
    Explanations

    tags with attributes

    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
     only
    -1.36
     that
    -1.18
     when
    -1.09
     Before
    -1.09
     these
    -1.08
    ........
    -1.06
     those
    -1.05
     Since
    -1.05
     before
    -1.02
     While
    -1.02
    POSITIVE LOGITS
     discuter
    1.38
     refroid
    1.28
    yyb
    1.20
    beginning
    1.17
     konkrét
    1.16
    Initializing
    1.14
    efter
    1.14
     varier
    1.14
    cheers
    1.13
    なのでしょう
    1.13
    Activations Density 0.008%

    No Known Activations