© 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-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 13970
    Prev
    Next
    INDEX
    Explanations

    prepositions indicating spatial or directional relationships

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

    Prepositions "on", "in", and "from"

    np_acts-logits-general · gemini-2.0-flash

    in phrases + locations/technical contexts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    bitField
    -0.36
    noopener
    -0.35
    awtextra
    -0.33
     correctes
    -0.33
    SourceChecksum
    -0.33
    abstractmethod
    -0.31
    Paraguay
    -0.31
     heißt
    -0.31
    ṁ
    -0.30
     lendemain
    -0.30
    POSITIVE LOGITS
     the
    0.61
     ddelwed
    0.59
     EconPapers
    0.58
    herited
    0.56
     various
    0.54
     a
    0.53
    UIControlState
    0.52
    ulemon
    0.51
    warnai
    0.51
    mastered
    0.51
    Activations Density 0.597%

    No Known Activations