© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 10276
    Prev
    Next
    INDEX
    Explanations

    someone giving an account of how people spent an evening

    oai_token-act-pair · gemini-2.0-flash

    conversation-related terms

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

    talking/chatting

    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
    ########.
    -1.10
     AttributeSet
    -0.72
    pexpr
    -0.66
    Datuak
    -0.65
    temming
    -0.60
    клопе
    -0.60
    __':
    -0.60
    addCriterion
    -0.59
    UpInside
    -0.59
     bağlantılar
    -0.59
    POSITIVE LOGITS
     talking
    1.38
     chatting
    1.34
     conversation
    1.34
     talk
    1.31
     chat
    1.29
     Talk
    1.21
    talk
    1.16
    talking
    1.16
     chatted
    1.15
    Talking
    1.13
    Activations Density 2.200%

    No Known Activations