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-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 12890
    Prev
    Next
    INDEX
    Explanations

    occurrences of variables, method calls, and data structures in programming languages

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
     שוליים
    -0.56
    KURZBESCHREIBUNG
    -0.54
     betweenstory
    -0.52
    ﹠
    -0.50
    oxo
    -0.50
     Ferns
    -0.50
    LikeLike
    -0.49
    Gambas
    -0.49
     кислота
    -0.48
    ckey
    -0.48
    POSITIVE LOGITS
    #+#
    0.42
    getDoctrine
    0.38
     Abonnez
    0.36
    autoreleasepool
    0.36
    InjectAttribute
    0.35
     indisponible
    0.35
     gynnwys
    0.35
    󠁴
    0.35
    TabStop
    0.32
     kiin
    0.30
    Activations Density 0.092%

    No Known Activations