© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 13095
    Prev
    Next
    INDEX
    Explanations

    statements related to ethical behavior and accountability.

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

    information related to nutrition, food fortification, and socioeconomic factors impacting public health.

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    the word "however" and similar transitional phrases that indicate contrast or qualification of a previous statement.

    oai_token-act-pair · claude-3-5-sonnet-20240620Triggered by @bot

    Followed by a comma or period

    np_acts-logits-general · gemini-2.0-flash

    however, course, said, notes

    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
    exitRule
    -0.70
     in
    -0.54
     as
    -0.51
     for
    -0.51
     despite
    -0.50
     while
    -0.50
     which
    -0.48
     according
    -0.47
     on
    -0.46
     with
    -0.46
    POSITIVE LOGITS
     Wiktionnaire
    0.81
    énario
    0.71
    lgari
    0.71
    MOQ
    0.70
    RUnlock
    0.65
    ruptedException
    0.65
     мәкал
    0.64
    expandindo
    0.64
    Lähteet
    0.62
    AsUp
    0.61
    Activations Density 0.386%

    No Known Activations