Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 64878
    Prev
    Next
    INDEX
    Explanations

    references to bait and related concepts in various contexts

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

    fishing bait or lures

    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
     EconPapers
    -0.75
    mybatisplus
    -0.75
     kasarigan
    -0.67
     للاسماء
    -0.65
     Infórmanos
    -0.63
    المكان
    -0.61
     continúas
    -0.60
    Географиясе
    -0.59
     nahilalakip
    -0.57
    tanleria
    -0.56
    POSITIVE LOGITS
     bait
    1.01
     Bait
    0.99
    Bait
    0.96
    bait
    0.82
     baits
    0.71
    饵
    0.58
     cebo
    0.52
    feed
    0.50
     marins
    0.43
     lure
    0.43
    Activations Density 0.003%

    No Known Activations