© 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 - 65k
    5. 16-GEMMASCOPE-RES-65K
    6. 28329
    Prev
    Next
    INDEX
    Explanations

    comparative phrases highlighting relationships or traits

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

    as ... as

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    context
    -0.49
     versus
    -0.47
     context
    -0.46
     Both
    -0.45
     вариант
    -0.45
    /
    -0.45
    ,
    -0.45
     group
    -0.44
     \
    -0.43
    mę
    -0.43
    POSITIVE LOGITS
    хьтан
    0.84
    BibitemShut
    0.81
    haustible
    0.79
    RectangleBorder
    0.78
    NUMX
    0.77
     HasFactory
    0.76
     חיצוניים
    0.75
    BagLayout
    0.75
     насељу
    0.73
    bootstrapcdn
    0.72
    Activations Density 0.086%

    No Known Activations