© 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. 215
    Prev
    Next
    INDEX
    Explanations

    expressions of understanding or confusion

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

    Preceding words like "understand," "sense," or "figured"

    np_acts-logits-general · gemini-2.0-flash

    understanding or confusion

    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
    CloseOperation
    -0.69
    writeFieldEnd
    -0.65
     Taktlose
    -0.65
    Itoa
    -0.63
    rxjs
    -0.63
     Normdatei
    -0.57
    titleMargin
    -0.57
    styleType
    -0.55
    /**
    -0.55
     BoxDecoration
    -0.55
    POSITIVE LOGITS
     understanding
    2.67
     understand
    2.47
     understood
    2.43
     Understanding
    2.41
    understanding
    2.34
    Understanding
    2.32
    understand
    2.21
     Understand
    2.17
     understands
    2.12
    Understand
    2.08
    Activations Density 0.469%

    No Known Activations