© 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. 20-GEMMASCOPE-RES-131K
    11. 61997
    Prev
    Next
    INDEX
    Explanations

    terms related to safety and prevention in hazardous situations

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

    fire related hazards and actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
     CanadaChoose
    -0.53
     Administrativna
    -0.47
     macam
    -0.43
    يكب
    -0.41
     braccia
    -0.41
     ligiloj
    -0.41
     jupe
    -0.40
     itulah
    -0.39
     ExecuteAsync
    -0.39
    Referanser
    -0.39
    POSITIVE LOGITS
    GEBURTSDATUM
    0.49
     Roskov
    0.45
     חיצוניים
    0.45
     conf
    0.44
    firestore
    0.43
     للاسماء
    0.42
     StringField
    0.39
    findpost
    0.38
     fire
    0.36
    VIAF
    0.35
    Activations Density 0.001%

    No Known Activations