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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 8920
    Prev
    Next
    INDEX
    Explanations

    specific structures and relationships in sentences, particularly focusing on the use of conjunctions and the context of various phrases

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

    after function words ("of", "to", "for")

    np_acts-logits-general · gemini-2.0-flash

    of the

    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
     justiça
    -0.31
    ">@
    -0.27
     consciência
    -0.25
     prêtre
    -0.25
     süre
    -0.24
    setVerticalGroup
    -0.24
    melada
    -0.24
     Schweitzer
    -0.24
     cobertura
    -0.24
    uestamente
    -0.24
    POSITIVE LOGITS
     kaarangay
    0.77
    存于互联网档案馆
    0.75
     beginnetje
    0.74
    <unused28>
    0.74
    <unused41>
    0.73
    <unused14>
    0.73
    <pad>
    0.73
    <unused17>
    0.73
    <unused16>
    0.73
    <unused8>
    0.73
    Activations Density 1.103%

    No Known Activations