© 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. 5917
    Prev
    Next
    INDEX
    Explanations

    structure-related elements in technical documents or code

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

    Tokens followed by "right" or "Right"

    np_acts-logits-general · gemini-2.0-flash

    right and directional modifiers

    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
     moindre
    -0.66
     мень
    -0.65
     lesser
    -0.64
     smaller
    -0.63
    Minor
    -0.59
    menor
    -0.58
     menor
    -0.56
    smaller
    -0.55
    ◄
    -0.54
    minlength
    -0.54
    POSITIVE LOGITS
    right
    0.92
     right
    0.90
     righ
    0.85
    Right
    0.84
     Right
    0.83
    getRight
    0.81
     RIGHT
    0.80
    EndProject
    0.80
    RIGHT
    0.80
     للاسماء
    0.78
    Activations Density 0.398%

    No Known Activations