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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 4893
    Prev
    Next
    INDEX
    Explanations

    references to violence and threats to safety

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

    Attacking, targeting, or harming others

    np_acts-logits-general · gemini-2.0-flash

    violence against innocents

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    ArgsConstructor
    -0.62
    Fprintf
    -0.54
    יצוני
    -0.53
    strerror
    -0.48
     igény
    -0.46
    toHave
    -0.46
     tác
    -0.45
     FontWeight
    -0.45
     számára
    -0.44
     gestos
    -0.44
    POSITIVE LOGITS
     unsuspecting
    1.35
     innocent
    1.20
     defen
    1.07
    innocent
    1.00
     inocente
    0.94
     unprotected
    0.91
     indiscrimin
    0.90
     helpless
    0.90
     innoc
    0.88
     hapless
    0.87
    Activations Density 0.488%

    No Known Activations