© 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. 31-GEMMASCOPE-RES-16K
    6. 8731
    Prev
    Next
    INDEX
    Explanations

    instances of verbs and phrases indicating action or decision-making

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

    the phrase "doing so" as a substitute for repeating a previously mentioned action.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    findpost
    -0.61
    OrWhiteSpace
    -0.58
     Chwiliwch
    -0.54
     يتيمه
    -0.52
    BagConstraints
    -0.48
     بلکه
    -0.47
    EndGlobalSection
    -0.47
    enumii
    -0.45
     nahilalakip
    -0.45
     useAppContext
    -0.44
    POSITIVE LOGITS
     so
    3.08
     так
    1.77
     så
    1.75
    so
    1.62
    如此
    1.54
     così
    1.52
     begitu
    1.45
     כך
    1.38
     niin
    1.34
    So
    1.29
    Activations Density 0.378%

    No Known Activations