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. 2698
    Prev
    Next
    INDEX
    Explanations

    code-related syntax, particularly method definitions and return statements

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

    Java code snippets / programming language elements

    np_acts-logits-general · gemini-2.0-flash

    return this, self

    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
    Erreferentziak
    -0.56
    Климат
    -0.49
    loadAll
    -0.45
    liesslich
    -0.45
    anso
    -0.44
    ########.
    -0.44
    Ран
    -0.43
     فريبيس
    -0.43
     człowieka
    -0.43
    ाल
    -0.42
    POSITIVE LOGITS
     this
    1.27
    this
    1.20
     этого
    0.67
     THIS
    0.67
     dieses
    0.67
    super
    0.66
     initComponents
    0.65
     цього
    0.65
     CreateTagHelper
    0.63
     questo
    0.62
    Activations Density 0.164%

    No Known Activations