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. 10699
    Prev
    Next
    INDEX
    Explanations

    technical elements related to formatting and coding standards in documents

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

    Comes before font-variant:small-caps

    np_acts-logits-general · gemini-2.0-flash

    small-caps, standard, int

    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
     '\\;'
    -1.14
    InjectAttribute
    -1.12
    rungsseite
    -1.11
    expandindo
    -1.08
     étoient
    -1.07
    ագրություններ
    -1.05
    NameInMap
    -1.04
    QMetaType
    -1.01
     AssemblyCulture
    -0.98
     ainfi
    -0.97
    POSITIVE LOGITS
    </strong>
    0.56
    <unused62>
    0.55
    <unused61>
    0.51
    </u>
    0.51
    </b>
    0.49
    <unused63>
    0.49
    <unused60>
    0.47
    =[];
    
    0.44
    同样
    0.41
    xhr
    0.40
    Activations Density 1.195%

    No Known Activations