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 - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 8295
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions and inequalities involving variables and relationships

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

    $1 \le i \le$

    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-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    ロウィン
    -0.75
     Geſch
    -0.73
    enablog
    -0.72
     Geiſt
    -0.71
    niſſe
    -0.71
     geſch
    -0.71
    <unused8>
    -0.71
    <unused14>
    -0.71
    <pad>
    -0.70
    <unused41>
    -0.70
    POSITIVE LOGITS
    1
    0.34
    L
    0.32
    P
    0.31
    Number
    0.30
    B
    0.30
    ostringstream
    0.29
    一
    0.28
    number
    0.28
    U
    0.28
    V
    0.28
    Activations Density 0.064%

    No Known Activations