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. 14511
    Prev
    Next
    INDEX
    Explanations

    sentences that include conditional or modal verbs and file access terminology

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

    German and Japanese text

    np_acts-logits-general · gemini-2.0-flash

    German and Japanese phrases

    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
     ikkje
    -0.68
     ikke
    -0.63
     שוליים
    -0.54
     välja
    -0.52
     ofta
    -0.49
     faktiskt
    -0.48
    invokeLater
    -0.47
    lamic
    -0.47
     fortfarande
    -0.47
     alun
    -0.47
    POSITIVE LOGITS
     is
    0.93
    는
    0.91
    は
    0.76
    은
    0.73
    서는
    0.72
     hat
    0.71
     shall
    0.68
    이는
    0.67
     は
    0.66
    지는
    0.64
    Activations Density 0.047%

    No Known Activations