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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 9726
    Prev
    Next
    INDEX
    Explanations

    elements related to news articles and reviews

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

    Follows punctuation or special characters

    np_acts-logits-general · gemini-2.0-flash

    applying for

    np_acts-logits-general · gemini-2.5-flash-lite

    indicators of document structure—headlines/section headers, metadata lines (dates/bylines), separators, and markup tags.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 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
     ivelany
    -0.77
    TagMode
    -0.72
    afficheront
    -0.68
    ➌
    -0.66
     '{@
    -0.66
     تضيفلها
    -0.65
     صوتيه
    -0.64
     uLocal
    -0.63
    ----</
    -0.63
    цездатний
    -0.63
    POSITIVE LOGITS
    自
    0.32
     An
    0.31
    </b>
    0.28
    あり
    0.28
    “
    0.28
    An
    0.27
    The
    0.27
     joindre
    0.27
     hat
    0.27
    OK
    0.27
    Activations Density 0.566%

    No Known Activations