Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 11163
    Prev
    Next
    INDEX
    Explanations

    legal and formal references within documents, including specific designations and citations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    übung
    -0.32
     оказалось
    -0.28
     tardes
    -0.26
    ConstraintMaker
    -0.26
     displays
    -0.26
     crees
    -0.26
     résist
    -0.26
    发表于
    -0.25
     הלא
    -0.24
     Widerstand
    -0.24
    POSITIVE LOGITS
    GTCX
    0.73
    ANTLR
    0.66
     hereafter
    0.66
     hereinafter
    0.65
    ſchaft
    0.63
    AddHtmlAttribute
    0.62
    таратура
    0.60
    ьаж
    0.60
     Wikimedijinoj
    0.60
    iſten
    0.60
    Activations Density 0.540%

    No Known Activations