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. 15971
    Prev
    Next
    INDEX
    Explanations

    references to online interactions, events, and educational sessions

    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
     urbanos
    -0.38
     permukaan
    -0.36
     manteau
    -0.34
     anggaran
    -0.33
     lema
    -0.32
    פט
    -0.32
    保养
    -0.32
     charbon
    -0.31
     bruit
    -0.31
     voet
    -0.31
    POSITIVE LOGITS
     virtual
    0.72
     Zoom
    0.68
    Zoom
    0.67
     livestream
    0.67
     webinars
    0.66
    Virtual
    0.66
    brtc
    0.65
     virtu
    0.65
     webinar
    0.65
     Virtual
    0.64
    Activations Density 0.124%

    No Known Activations