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-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 2063
    Prev
    Next
    INDEX
    Explanations

    technical details and capabilities of monitoring systems and devices

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 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
     entweder
    -0.62
     pinulongan
    -0.61
     Decke
    -0.61
     Though
    -0.60
    either
    -0.60
    सन्दर्भ
    -0.59
     postIndex
    -0.59
    Americas
    -0.56
    }{*}{}
    -0.56
    ditional
    -0.55
    POSITIVE LOGITS
    StoryboardSegue
    0.78
    UnusedPrivate
    0.63
    onViewCreated
    0.61
    jsonwebtoken
    0.56
     JpaRepository
    0.53
    ::::::::
    0.53
     conséquent
    0.53
    UrlResolution
    0.52
     the
    0.52
    TagMode
    0.52
    Activations Density 0.153%

    No Known Activations