© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 2-GEMMASCOPE-RES-16K
    6. 9459
    Prev
    Next
    INDEX
    Explanations

    contexts related to the term "Quad," particularly in architectural and campus settings

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67
    Prompts (Dashboard)
    24,576 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
     feroit
    -0.45
    TokenNameDOT
    -0.43
     ums
    -0.42
     Kec
    -0.42
     Tasse
    -0.42
    PropTypes
    -0.42
    getKeyCode
    -0.42
     Schur
    -0.41
    rhe
    -0.41
     OMITBAD
    -0.41
    POSITIVE LOGITS
     quad
    1.08
     Quad
    1.03
    Quad
    0.96
     quads
    0.77
     quadru
    0.75
     duo
    0.70
    QUAD
    0.68
     mono
    0.67
    quad
    0.67
     bi
    0.65
    Activations Density 0.640%

    No Known Activations