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 - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 14541
    Prev
    Next
    INDEX
    Explanations

    sequences of underscores and symbols indicating programming or technical syntax

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

    Comes after "_" in code or data

    np_acts-logits-general · gemini-2.0-flash

    function/method calls, pauses, or group names

    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
     Inscrivez
    -0.51
     avoient
    -0.46
     olvido
    -0.45
     amanhã
    -0.45
     desconocido
    -0.44
     salvaje
    -0.43
     alrededores
    -0.43
     enfermo
    -0.43
     antiguo
    -0.42
     Könige
    -0.42
    POSITIVE LOGITS
    Rhestr
    0.69
    twimg
    0.66
    riwal
    0.65
    atorship
    0.64
    帖最后由
    0.62
     تكبرها
    0.62
    PyExc
    0.61
    postolic
    0.60
     nonatomic
    0.60
    liferay
    0.60
    Activations Density 0.800%

    No Known Activations