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. 9121
    Prev
    Next
    INDEX
    Explanations

    code elements and structure related to graphical user interface (GUI) components

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

    Position or direction (left, lower, top)

    np_acts-logits-general · gemini-2.0-flash

    layout and code elements

    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
    +#+#
    -0.74
    parsedMessage
    -0.68
     kaarangay
    -0.65
    ViewFeatures
    -0.63
     ExecuteAsync
    -0.59
    adaptiveStyles
    -0.58
    localctx
    -0.58
    tagHelperRunner
    -0.55
    OGND
    -0.55
     Autorizaciones
    -0.54
    POSITIVE LOGITS
    новништво
    0.47
     Bekasi
    0.43
     poffe
    0.41
     Majefty
    0.39
     ſp
    0.39
     preſ
    0.38
     Chriftian
    0.36
    ┳
    0.36
    upvote
    0.36
     camicia
    0.35
    Activations Density 0.606%

    No Known Activations