© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 56186
    Prev
    Next
    INDEX
    Explanations

    occurrences of specific genes, treatments, or biological processes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    books and measurements

    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
    Not in Any Lists

    No Comments

    Negative Logits
     adelantado
    -0.37
    WebElementEntity
    -0.36
    ConfigureAwait
    -0.36
    「
    -0.35
    ViewImports
    -0.35
    /**
    
    -0.34
     unfounded
    -0.34
    astéroïdes
    -0.34
     tuin
    -0.33
     Paglinawan
    -0.33
    POSITIVE LOGITS
     myſelf
    0.66
     Cæsar
    0.62
     Efq
    0.62
    twimg
    0.61
     Anſ
    0.60
     ſind
    0.60
     Theſe
    0.59
     protoimpl
    0.57
    ſammen
    0.57
     Traité
    0.57
    Activations Density 0.501%

    No Known Activations