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. 31-GEMMASCOPE-RES-16K
    6. 15610
    Prev
    Next
    INDEX
    Explanations

    key components and features related to computer hardware and their compatibility

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

    Asrock P4VM8 motherboard and hard disks

    np_acts-logits-general · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    cu
    -0.29
    stage
    -0.27
    static
    -0.26
    <strong>
    -0.26
     $\
    -0.25
     stage
    -0.25
     brauche
    -0.25
    舞台
    -0.25
    \_
    -0.24
    on
    -0.24
    POSITIVE LOGITS
    rungsseite
    0.84
     surla
    0.81
     AssemblyCulture
    0.69
     Infórmanos
    0.68
    outSlope
    0.66
    MLLoader
    0.63
     otomatig
    0.63
    StoryboardSegue
    0.63
    inSlope
    0.62
    transQ
    0.62
    Activations Density 0.031%

    No Known Activations