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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 7920
    Prev
    Next
    INDEX
    Explanations

    will cause strong positive effect

    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_34/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
    helps
    -0.98
     PUSTAKA
    -0.88
    helped
    -0.87
    iffen
    -0.84
     helpful
    -0.84
     valign
    -0.82
     membantu
    -0.82
    atshirt
    -0.82
     ragazze
    -0.82
    wiście
    -0.81
    POSITIVE LOGITS
     winning
    1.61
     tug
    1.56
     win
    1.53
     capture
    1.52
     leave
    1.45
     captured
    1.44
     stole
    1.41
     mesmer
    1.41
     wins
    1.34
     hypno
    1.30
    Activations Density 0.046%

    No Known Activations