Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 11808
    Prev
    Next
    INDEX
    Explanations

    hearing information

    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
    i
    -2.86
    m
    -2.45
    c
    -2.39
    g
    -2.30
    when
    -2.19
    which
    -2.19
     We
    -2.17
    s
    -2.17
    ised
    -2.08
    ஂ
    -2.08
    POSITIVE LOGITS
     the
    2.38
    
    2.27
     firsthand
    2.14
    xvii
    2.08
     plenty
    2.05
     großes
    2.03
    weakSelf
    2.02
    鑁
    1.99
     Nieuws
    1.99
     wykonane
    1.98
    Activations Density 0.015%

    No Known Activations