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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 65k
    9. 0-GEMMASCOPE-RES-65K
    11. 11089
    Prev
    Next
    INDEX
    Explanations

    the word "but" used to introduce a contrast or exception

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 0-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_0/width_65k/average_l0_73
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.0.hook_resid_post
    Hook Layer
    0
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Efq
    -1.24
     itſelf
    -1.21
    ſelf
    -1.20
     Theſe
    -1.20
     raiſ
    -1.17
     Jefus
    -1.14
     myſelf
    -1.13
     iſt
    -1.12
     ſche
    -1.09
     ―――――
    -1.08
    POSITIVE LOGITS
     but
    1.88
    but
    1.23
     BUT
    0.98
     pero
    0.98
     But
    0.94
     nhưng
    0.88
     maar
    0.86
     लेकिन
    0.85
     sondern
    0.85
     αλλά
    0.83
    Activations Density 0.207%

    No Known Activations