© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    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. 10448
    Prev
    Next
    INDEX
    Explanations

    elements related to programming functions, methods, and their parameters

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

    Code, names, and other "technical" language

    np_acts-logits-general · gemini-2.0-flash

    label for item typed

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     الحره
    -0.82
     صوتيه
    -0.72
    bootstrapcdn
    -0.67
    IntoConstraints
    -0.63
    GEBURTSDATUM
    -0.63
    <unused14>
    -0.63
    <unused74>
    -0.63
    <unused16>
    -0.62
    [@BOS@]
    -0.62
    <pad>
    -0.62
    POSITIVE LOGITS
     overras
    0.41
    chapper
    0.30
    nowe
    0.29
     saan
    0.29
    Where
    0.29
    principalColumn
    0.28
    Name
    0.28
     yeterli
    0.28
    лору
    0.27
    Parent
    0.27
    Activations Density 0.367%

    No Known Activations