Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 75158
    Prev
    Next
    INDEX
    Explanations

    concepts related to crowd-sourcing and community collaboration

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

    crowd-sourcing and collaboration

    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
    enderror
    -0.33
     dinding
    -0.31
     conversión
    -0.28
     bekannt
    -0.28
     preparación
    -0.27
     prefeito
    -0.27
    awancara
    -0.26
     posición
    -0.25
     olup
    -0.24
     she
    -0.24
    POSITIVE LOGITS
    <pad>
    0.68
    ๬
    0.68
    <unused52>
    0.67
    <unused42>
    0.67
    <unused68>
    0.67
    <unused74>
    0.67
    <unused41>
    0.67
    <unused11>
    0.67
    <unused8>
    0.67
    <unused3>
    0.67
    Activations Density 0.020%

    No Known Activations