© 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. 7066
    Prev
    Next
    INDEX
    Explanations

    information related to spells and abilities in a magical context

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

    followed by "of," "to," "in," "as," or "that"

    np_acts-logits-general · gemini-2.0-flash

    explaining concepts or reasons

    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
    ajara
    -0.63
    vosti
    -0.54
    ArrowToggle
    -0.52
    :✨
    -0.51
    ñera
    -0.50
    ENSATION
    -0.50
    ','
    -0.49
    ніципалі
    -0.49
    ","
    -0.49
    owie
    -0.48
    POSITIVE LOGITS
     typelib
    0.65
    IBOutlet
    0.63
    RectangleBorder
    0.61
    醐
    0.54
    stateProvider
    0.54
     sake
    0.54
     RoutedEventArgs
    0.52
     theirs
    0.51
    Personendaten
    0.51
     OnClickListener
    0.50
    Activations Density 0.373%

    No Known Activations