Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 18-GEMMASCOPE-RES-16K
    6. 1535
    Prev
    Next
    INDEX
    Explanations

    sections and references related to compliance and legal agreements

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

    "section" or words before "compliance" or "Appendix"

    np_acts-logits-general · gemini-2.0-flash

    compliance with license agreements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Portale
    -0.75
    بوابة
    -0.73
    AsUp
    -0.72
    FunctionFlags
    -0.70
    __*/
    -0.68
     estekak
    -0.68
    MessageOf
    -0.66
    +#+#
    -0.63
    VIAF
    -0.63
    osoba
    -0.61
    POSITIVE LOGITS
    section
    0.81
    Section
    0.56
     section
    0.54
     cardiaque
    0.53
     Section
    0.52
     secco
    0.52
     Bellini
    0.51
    SequentialGroup
    0.51
    {
    0.50
    *{
    0.49
    Activations Density 0.001%

    No Known Activations