© 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. 117152
    Prev
    Next
    INDEX
    Explanations

    key components and attributes related to light bulbs and lamps

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

    types of lamps and bulbs

    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
    IRQn
    -0.52
     Rumuni
    -0.46
     referrerpolicy
    -0.41
    pragma
    -0.41
     ویکی‌پدیای
    -0.41
     tracking
    -0.41
     Tracking
    -0.40
     final
    -0.40
    iyaki
    -0.40
     tracked
    -0.40
    POSITIVE LOGITS
     bulb
    0.88
    bulb
    0.81
     Bulb
    0.79
    Bulb
    0.79
     bulbs
    0.77
     Bulbs
    0.73
    andescent
    0.72
     lampe
    0.66
     lâmpada
    0.65
     lamp
    0.62
    Activations Density 0.009%

    No Known Activations