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. 2295
    Prev
    Next
    INDEX
    Explanations

    phrases related to the processing and submission of forms and questionnaires

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

    Completing or filling out forms/questionnaires

    np_acts-logits-general · gemini-2.0-flash

    filling out forms

    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
    !*\
    -0.67
    ArrowToggle
    -0.66
    </tfoot>
    -0.60
    +:+
    -0.59
    tvguidetime
    -0.58
     Taktlose
    -0.58
    Rüyada
    -0.58
    illeur
    -0.58
     ExecuteAsync
    -0.57
    MemoryWarning
    -0.56
    POSITIVE LOGITS
     filling
    1.47
     fill
    1.44
     Fill
    1.35
     forms
    1.35
     form
    1.34
     Filling
    1.32
     filled
    1.31
     questionnaire
    1.28
    forms
    1.25
    Fill
    1.24
    Activations Density 0.366%

    No Known Activations