© 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. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 4642
    Prev
    Next
    INDEX
    Explanations

    words associated with eating and drinking

    oai_token-act-pair · gemini-2.0-flash

    prefixes related to food or drink

    np_token-act-pair-logits · gpt-4o-mini

    cup or shop

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    principalColumn
    -0.71
    LabelTagHelper
    -0.63
    KURZBESCHREIBUNG
    -0.61
    UnsafeEnabled
    -0.60
     AssemblyCompany
    -0.60
     <>",
    -0.57
    polation
    -0.56
    ɵɵelementStart
    -0.55
    becher
    -0.55
    ParallelGroup
    -0.55
    POSITIVE LOGITS
    ma
    0.51
     checking
    0.51
    du
    0.50
     éti
    0.50
    ++){
    0.50
    id
    0.49
    bard
    0.47
    })`
    0.47
    mas
    0.46
    nos
    0.46
    Activations Density 2.120%

    No Known Activations