© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 11489
    Prev
    Next
    INDEX
    Explanations

    descriptive imagery related to natural landscapes and water features

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

    Water flow in streams or rivers

    np_acts-logits-general · gemini-2.0-flash

    stream, river, flow

    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
     للمعارف
    -0.98
     المعيارى
    -0.69
    깥
    -0.67
    Билгалдахарш
    -0.65
    Демографія
    -0.64
    Tembelea
    -0.64
    numerusform
    -0.63
    BeginContext
    -0.61
     imageNamed
    -0.61
    contentLoaded
    -0.61
    POSITIVE LOGITS
     stream
    1.46
     river
    1.45
     rivers
    1.23
     streams
    1.23
     tributaries
    1.21
     Stream
    1.20
     downstream
    1.18
     tributary
    1.17
    stream
    1.15
     flow
    1.15
    Activations Density 0.190%

    No Known Activations