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. 2275
    Prev
    Next
    INDEX
    Explanations

    terms related to legal and corporate accountability

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

    followed by modal verbs

    np_acts-logits-general · gemini-2.0-flash

    obligation, permission, or advice

    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
     <<<<<<<<<<<<<<
    -0.76
    URLException
    -0.71
     архивлан
    -0.69
    Collegamenti
    -0.67
    moveToFirst
    -0.65
    lapi
    -0.65
    addCriterion
    -0.65
    PerformLayout
    -0.63
    inerary
    -0.62
    Lähteet
    -0.61
    POSITIVE LOGITS
     should
    0.86
     may
    0.67
     bør
    0.66
     please
    0.66
     must
    0.62
     can
    0.60
    should
    0.60
     möge
    0.54
     gebeten
    0.54
     wishing
    0.54
    Activations Density 0.237%

    No Known Activations