© 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-2-2B
    4. Residual Stream - 16k
    5. 8-GEMMASCOPE-RES-16K
    6. 4223
    Prev
    Next
    INDEX
    Explanations

    references to groups of three or combinations of three

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 8-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_8/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    circledR
    -0.54
     many
    -0.53
    zsche
    -0.51
    nostic
    -0.51
     Derecho
    -0.51
    stånd
    -0.51
     szóci
    -0.50
    tens
    -0.49
    นวน
    -0.49
    binant
    -0.48
    POSITIVE LOGITS
     trio
    1.25
     triad
    1.13
     Trio
    1.11
    three
    1.08
    THREE
    1.06
     triplets
    1.02
    三
    1.02
     trilogy
    1.01
     three
    1.01
     triplet
    1.01
    Activations Density 0.753%

    No Known Activations