© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 37955
    Prev
    Next
    INDEX
    Explanations

    references to attention and focus on actions or resources

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

    focus attention on

    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
     coster
    -0.40
     gall
    -0.39
    Admission
    -0.36
    admission
    -0.35
    zeichnet
    -0.35
     grosses
    -0.35
     impré
    -0.35
     Admission
    -0.35
     patente
    -0.35
     occafion
    -0.34
    POSITIVE LOGITS
     focus
    0.73
     attention
    0.69
     focused
    0.68
     focussed
    0.67
    focus
    0.63
     concentrated
    0.60
     focuses
    0.59
     focussing
    0.59
    focused
    0.58
     concentrating
    0.57
    Activations Density 0.027%

    No Known Activations