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-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 1548
    Prev
    Next
    INDEX
    Explanations

    expressions of personal giving or philanthropic actions

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

    Past tense verbs

    np_acts-logits-general · gemini-2.0-flash

    personal past states

    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
     are
    -0.66
    ಿದೆ
    -0.63
    렵
    -0.58
     Recently
    -0.58
    Says
    -0.56
     will
    -0.56
     bislang
    -0.56
     currently
    -0.56
    Хочу
    -0.56
    になっています
    -0.55
    POSITIVE LOGITS
     was
    2.22
     were
    1.78
     had
    1.68
     wasn
    1.68
    was
    1.63
     seemed
    1.52
     didn
    1.50
    Was
    1.44
     knew
    1.44
     weren
    1.43
    Activations Density 7.055%

    No Known Activations