© 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. 2-GEMMASCOPE-RES-16K
    6. 821
    Prev
    Next
    INDEX
    Explanations

    phrases that emphasize quantity or significance related to "a" or "an."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 2-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    InputBorder
    -0.77
    twimg
    -0.74
     postIndex
    -0.72
    IntoConstraints
    -0.66
    ElementException
    -0.66
    currentColor
    -0.61
     manik
    -0.58
     Krakowie
    -0.58
     nonUne
    -0.58
    RenderAtEndOf
    -0.58
    POSITIVE LOGITS
    作為
    0.71
     як
    0.69
     как
    0.68
     jako
    0.68
    作为一个
    0.67
    作为
    0.66
     sebagai
    0.61
     kaip
    0.60
     as
    0.58
    Sebagai
    0.54
    Activations Density 0.473%

    No Known Activations