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. 10934
    Prev
    Next
    INDEX
    Explanations

    curly braces and other related syntactical elements in structured data formats

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

    Code snippets or configuration blocks

    np_acts-logits-general · gemini-2.0-flash

    references and code blocks

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    -
    -0.64
    ので
    -0.58
    .
    -0.55
    ,
    -0.54
     μέ
    -0.53
     Pal
    -0.49
     AT
    -0.48
     K
    -0.47
    yar
    -0.47
     uma
    -0.46
    POSITIVE LOGITS
     purpoſe
    1.05
     myſelf
    1.05
    ⓧ
    1.02
     themſelves
    1.01
     feroit
    1.01
    Disliked
    1.01
     pleaſure
    1.00
     auffi
    1.00
     cauſe
    0.99
     reaſon
    0.98
    Activations Density 0.715%

    No Known Activations