© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 27-GEMMASCOPE-RES-16K
    11. 12900
    Prev
    Next
    INDEX
    Explanations

    code involving incrementing and decrementing operations

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 27-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_27/width_16k/average_l0_118
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.27.hook_resid_post
    Hook Layer
    27
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    ⟬
    -1.01
    RenderAtEndOf
    -0.85
    
    -0.84
    tagHelperRunner
    -0.81
     ProtoMessage
    -0.80
     surla
    -0.80
     Taktlose
    -0.77
    Tembelea
    -0.76
     AspNetCore
    -0.75
     Normdatei
    -0.75
    POSITIVE LOGITS
     advance
    0.39
     nasled
    0.39
     avan
    0.38
     passed
    0.33
     further
    0.33
     next
    0.33
     weiter
    0.33
     avancer
    0.33
     cereales
    0.32
     advanced
    0.31
    Activations Density 0.105%

    No Known Activations