© 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. Gemma-2-2B
    3. 0-GEMMASCOPE-TRANSCODER-16K
    4. 6381
    Prev
    Next
    INDEX
    Explanations

    negative situations or experiences

    oai_token-act-pair · gemini-2.0-flash

    threatened

    np_max-act · gemini-2.0-flash

    threatened

    np_max-act-logits · gpt-5-miniTriggered by @chenshw0109
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_0/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.0.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
    DockStyle
    -1.38
    __":
    
    -1.34
    脚注の使い方
    -1.31
     незавершена
    -1.30
     estekak
    -1.29
    UnusedPrivate
    -1.28
     <=",
    -1.28
     nahilalakip
    -1.23
     HasFactory
    -1.23
     resourceCulture
    -1.22
    POSITIVE LOGITS
     of
    0.88
     (
    0.84
     to
    0.75
     or
    0.74
     "
    0.71
     
    0.70
     $
    0.68
     individual
    0.67
     I
    0.67
     with
    0.66
    Activations Density 2.313%

    No Known Activations