© 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. 11994
    Prev
    Next
    INDEX
    Explanations

    words that often appear in conversational or informal settings

    oai_token-act-pair · gemini-2.0-flash

    words indicating conjunction or relative clauses

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

    "that" in other languages

    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
    quels
    -0.72
    ::*;
    -0.46
    ندگان
    -0.44
    Werdegang
    -0.44
    PYX
    -0.43
    ituri
    -0.43
    ManagedBean
    -0.43
    IEU
    -0.43
    quelles
    -0.42
    chanel
    -0.42
    POSITIVE LOGITS
     that
    3.25
    that
    2.05
     bahwa
    1.88
     rằng
    1.75
     dass
    1.63
     że
    1.57
     kwamba
    1.45
     että
    1.41
     bahawa
    1.41
     ότι
    1.40
    Activations Density 20.471%

    No Known Activations