© 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-IT
    7. Residual Stream - 131k
    9. 31-GEMMASCOPE-RES-131K
    11. 17070
    Prev
    Next
    INDEX
    Explanations

    mathematical alignment and formatting structures in the text

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot

    LaTeX alignment endings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
     pinulongan
    -0.41
    ंदीखरीदारी
    -0.35
     fri
    -0.34
    getStringExtra
    -0.33
     dets
    -0.31
     rep
    -0.31
    ぼちゃ
    -0.31
    Gambas
    -0.31
     ori
    -0.30
     orange
    -0.30
    POSITIVE LOGITS
     HasFactory
    0.68
    ագրություններ
    0.63
     kasarigan
    0.59
    UrlResolution
    0.58
    verwijspagina
    0.54
     Schluß
    0.54
    cemment
    0.54
    equation
    0.54
     endfor
    0.53
    المراجع
    0.53
    Activations Density 0.022%

    No Known Activations