© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 9045
    Prev
    Next
    INDEX
    Explanations

    function definitions and their constructors in code snippets

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

    Tokens starting with "__"

    np_acts-logits-general · gemini-2.0-flash

    double underscore identifiers

    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
    Lund
    -0.59
    .”.
    -0.53
    NTB
    -0.53
     Tind
    -0.52
    !”.
    -0.51
    itv
    -0.51
    )”.
    -0.51
    .},
    -0.51
    antd
    -0.50
    AndEndTag
    -0.50
    POSITIVE LOGITS
     __
    1.09
    (__
    1.03
    .__
    0.98
     "__
    0.94
    ->__
    0.90
     (__
    0.90
    __
    0.88
    ("__
    0.86
     *__
    0.85
     '__
    0.84
    Activations Density 0.088%

    No Known Activations