© 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. 1817
    Prev
    Next
    INDEX
    Explanations

    specific sections or portions within a document that discuss claims or details about legal matters

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

    Parts, segments, portions, or phases

    np_acts-logits-general · gemini-2.0-flash

    parts, portions, phases, segments

    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
    
    -0.42
    Datuak
    -0.39
     among
    -0.34
    +:+
    -0.32
    sendKeys
    -0.31
    łaj
    -0.31
    ridae
    -0.31
    ⓘ
    -0.31
     possible
    -0.31
     mahd
    -0.31
    POSITIVE LOGITS
     portion
    0.90
     part
    0.85
     Portion
    0.85
    Part
    0.82
    portion
    0.82
    part
    0.79
     część
    0.77
     část
    0.77
     Bagian
    0.76
    Portion
    0.76
    Activations Density 0.953%

    No Known Activations