© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 17-GEMMASCOPE-RES-16K
    11. 1818
    Prev
    Next
    INDEX
    Explanations

    instances of the word "say" combined with expressions of understatement or strong opinions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 17-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_17/width_16k/average_l0_77
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.17.hook_resid_post
    Hook Layer
    17
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    StoryboardSegue
    -0.62
     driver
    -0.58
    MessageOf
    -0.56
    حياته
    -0.55
    AndEndTag
    -0.55
    posedge
    -0.55
     Púb
    -0.54
    حياتها
    -0.53
     drivers
    -0.51
    SourceChecksum
    -0.51
    POSITIVE LOGITS
     understatement
    0.65
     indeed
    0.59
    ล์
    0.57
     Meksiku
    0.56
     least
    0.53
     downright
    0.53
     suffice
    0.52
    least
    0.51
    dega
    0.51
    ächlich
    0.50
    Activations Density 0.154%

    No Known Activations