© 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
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 14529
    Prev
    Next
    INDEX
    Explanations

    references to different versions or implementations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    dflare
    -0.50
    NOTIFY
    -0.48
     administratif
    -0.45
    ContextHolder
    -0.45
     Schmel
    -0.43
     publicité
    -0.42
     charité
    -0.42
     Kna
    -0.40
     SALUD
    -0.40
    Primary
    -0.40
    POSITIVE LOGITS
     version
    1.55
     Version
    1.38
    version
    1.30
     versions
    1.29
     VERSION
    1.27
    Version
    1.25
     Versions
    1.14
     versione
    1.09
     getVersion
    1.05
     versión
    1.03
    Activations Density 0.150%

    No Known Activations