© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 899
    Prev
    Next
    INDEX
    Explanations

    scientific terms and specific classifications related to biological or chemical entities

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

    ending with "cler", "dic", or "duk"

    np_acts-logits-general · gemini-2.0-flash

    sclerosponges, aedicula, ddukbokkie, actorale, Hmaid, coaptation, isotype

    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
    -
    -0.51
    <bos>
    -0.49
    o
    -0.48
     fuss
    -0.47
    [toxicity=0]
    -0.45
     merit
    -0.44
    n
    -0.43
    t
    -0.42
    arakhand
    -0.41
     delight
    -0.41
    POSITIVE LOGITS
    balleur
    0.60
    majánló
    0.60
    ourtney
    0.60
     ſche
    0.59
     Verſ
    0.59
    ágenes
    0.57
     auffi
    0.57
     Sénat
    0.57
    ientras
    0.56
    actéristi
    0.56
    Activations Density 1.031%

    No Known Activations