© 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. 11970
    Prev
    Next
    INDEX
    Explanations

    qualifying adverbs and conjunctions used to emphasize the importance or extent of a following statement.

    oai_token-act-pair · gemini-2.0-flash

    adverbs and adjective endings

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

    ending with "ly"

    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
     récentes
    -0.71
    seamnă
    -0.70
     sociala
    -0.70
     récents
    -0.70
     particulières
    -0.67
    atalos
    -0.65
     bedoeld
    -0.65
     chré
    -0.65
     대한
    -0.65
     avoient
    -0.63
    POSITIVE LOGITS
    ingly
    1.27
    antly
    1.25
    ishly
    1.24
    ently
    1.23
    lessly
    1.22
    matically
    1.22
    ously
    1.21
    ically
    1.19
    ly
    1.19
    ably
    1.16
    Activations Density 3.880%

    No Known Activations