© 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 - 65k
    9. 20-GEMMASCOPE-RES-65K
    11. 135
    Prev
    Next
    INDEX
    Explanations

    phrases expressing subjective evaluations and descriptions

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

    expressions of desire or aspiration

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

    best way to describe

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
    SourceChecksum
    -0.66
    GEBURTSDATUM
    -0.47
    Buildable
    -0.45
    InjectAttribute
    -0.44
    イク
    -0.43
    machung
    -0.43
    /-/
    -0.43
    Lähteet
    -0.43
    Granted
    -0.43
    pektor
    -0.43
    POSITIVE LOGITS
     describe
    2.08
     description
    2.03
     describing
    1.96
     describes
    1.93
     descriptions
    1.80
    describing
    1.73
     Descri
    1.72
    description
    1.70
     descri
    1.69
     descrip
    1.69
    Activations Density 0.099%

    No Known Activations