© 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. 22-GEMMASCOPE-RES-16K
    6. 9740
    Prev
    Next
    INDEX
    Explanations

    technical terminology and mathematical concepts related to set theory and topology

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

    Mathematical sets, subsets, and cones

    np_acts-logits-general · gemini-2.0-flash

    sets, subsets, and emptiness

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    uxxxx
    -0.61
    rungsseite
    -0.54
    errHandler
    -0.48
     rigged
    -0.46
     pulpit
    -0.46
     CGPoint
    -0.46
     strdup
    -0.46
    GGT
    -0.46
     atof
    -0.45
    出版年
    -0.45
    POSITIVE LOGITS
     subset
    1.02
     subsets
    0.94
    subset
    0.93
     disambiguazione
    0.84
    subseteq
    0.82
     containment
    0.79
    ⊆
    0.77
    Subset
    0.75
     муніципалі
    0.73
     contained
    0.73
    Activations Density 1.689%

    No Known Activations