© 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. Qwen3-4B
    5. 19-TRANSCODER-HP
    7. 4957
    Prev
    Next
    INDEX
    Explanations

    scraped data, machine data

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    amble
    -0.28
    æ¯ıä¸Ģ次
    -0.27
    )||(
    -0.26
    HEN
    -0.26
    spar
    -0.25
    &&(
    -0.25
    éĶŃ
    -0.25
    åıijå±ķ模å¼ı
    -0.24
    amina
    -0.24
    -human
    -0.24
    POSITIVE LOGITS
    设计
    0.30
    è¿ĻæĿ¡
    0.28
    å¼Ģ设
    0.27
    åĪĩ
    0.26
     design
    0.25
    https
    0.25
     #
    0.25
     shel
    0.25
    åĿ¦
    0.25
    å¹»
    0.25
    Activations Density 0.004%

    No Known Activations