© 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. 4625
    Prev
    Next
    INDEX
    Explanations

    phrases related to associations and collaborations within organizations

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

    or phrases followed by citations or specifications

    np_acts-logits-general · gemini-2.0-flash

    clause boundaries

    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
    IntoConstraints
    -0.75
     OMITBAD
    -0.71
     createState
    -0.66
     MainAxisSize
    -0.64
     Administrativna
    -0.63
     nahilalakip
    -0.62
    aarrggbb
    -0.60
     initComponents
    -0.59
    <unused52>
    -0.58
    <unused74>
    -0.58
    POSITIVE LOGITS
    sproz
    0.35
    lab
    0.33
    which
    0.32
    Successfully
    0.32
    nezeu
    0.32
     which
    0.30
    angnya
    0.30
    (@"%@",
    0.30
    yonel
    0.29
    department
    0.29
    Activations Density 0.962%

    No Known Activations