© 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. 22-GEMMASCOPE-RES-16K
    6. 6376
    Prev
    Next
    INDEX
    Explanations

    references to wood and resin materials in a construction context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_22/width_16k/average_l0_123
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    =$?
    -0.51
     Soccerway
    -0.46
    Tikang
    -0.46
     häl
    -0.45
    tagext
    -0.44
    stylers
    -0.42
    stateMutability
    -0.40
    örm
    -0.40
     cuarzo
    -0.40
    RenderAtEndOf
    -0.40
    POSITIVE LOGITS
     wood
    1.29
     timber
    1.13
     Wood
    1.11
     Timber
    1.05
     woods
    1.04
     WOOD
    1.03
     Wooden
    1.02
     wooden
    1.02
     hardwood
    1.00
     woodworking
    1.00
    Activations Density 0.778%

    No Known Activations