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. 4898
    Prev
    Next
    INDEX
    Explanations

    components and terminology related to biomedical contexts

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

    ending in "i" followed by other letters

    np_acts-logits-general · gemini-2.0-flash

    word endings

    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
    ს
    -0.61
    nya
    -0.58
    pilar
    -0.54
    ن
    -0.49
    ם
    -0.49
    د
    -0.48
    source
    -0.47
    ness
    -0.47
    ند
    -0.47
     Feynman
    -0.46
    POSITIVE LOGITS
     JpaRepository
    0.60
    IBOutlet
    0.55
     Escri
    0.54
    řevě
    0.54
     AsyncTask
    0.53
    brainly
    0.53
    )|\
    0.53
    таратура
    0.52
    TagMode
    0.51
     cigarettes
    0.51
    Activations Density 0.276%

    No Known Activations