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. 14487
    Prev
    Next
    INDEX
    Explanations

    phrases related to decision-making and evaluations

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

    Follows pronouns and possessives

    np_acts-logits-general · gemini-2.0-flash

    different types

    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
    .
    -0.38
     beyond
    -0.31
    ,
    -0.29
    ↵↵
    -0.28
     during
    -0.27
    ?
    -0.27
    ;
    -0.26
    CR
    -0.26
     without
    -0.26
    This
    -0.25
    POSITIVE LOGITS
     kasarigan
    0.97
    ロウィン
    0.95
    hyrchwyd
    0.94
    ðsíða
    0.88
     pinulongan
    0.88
    ésultats
    0.88
    <unused80>
    0.87
    <unused51>
    0.86
    <unused41>
    0.86
    <unused21>
    0.86
    Activations Density 0.805%

    No Known Activations