© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 105561
    Prev
    Next
    INDEX
    Explanations

    penultimate, however, even, context introduction

    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_10/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
    蠖
    -2.66
    sudadera
    -2.55
    珖
    -2.48
    ﻂ
    -2.47
    congrat
    -2.39
    
    -2.39
    spoke
    -2.34
    namely
    -2.34
    grilled
    -2.30
     disposent
    -2.25
    POSITIVE LOGITS
    e
    3.16
     he
    2.88
    1
    2.84
     To
    2.59
     After
    2.58
     Three
    2.53
     We
    2.52
     Some
    2.50
    ischof
    2.44
    f
    2.44
    Activations Density 0.001%

    No Known Activations