© 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. 36-GEMMASCOPE-RES-16K
    6. 7923
    Prev
    Next
    INDEX
    Explanations

    elements related to vehicle features and materials

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.36.hook_resid_post
    Hook Layer
    36
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    GEBURTSDATUM
    -0.74
    èdia
    -0.56
    Erreferentziak
    -0.54
    WebElementEntity
    -0.53
    afficheront
    -0.52
    OGND
    -0.45
    xase
    -0.43
    endpush
    -0.43
    تقاوى
    -0.42
    Cyfarwyddwr
    -0.41
    POSITIVE LOGITS
     matte
    0.98
     matt
    0.95
     gloss
    0.92
     satin
    0.85
     glossy
    0.84
     Satin
    0.82
     Gloss
    0.81
    matte
    0.81
     Matte
    0.79
     finish
    0.78
    Activations Density 0.195%

    No Known Activations