© 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. Qwen3-4B
    5. 19-TRANSCODER-HP
    7. 7230
    Prev
    Next
    INDEX
    Explanations

    Code and foreign languages

    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
     respectfully
    -0.27
     needs
    -0.27
    Accept
    -0.25
    Sep
    -0.25
    欺è¯Ī
    -0.25
    ån
    -0.24
    åĢ®
    -0.24
    Keep
    -0.24
     acceptance
    -0.24
    _accept
    -0.24
    POSITIVE LOGITS
    交代
    0.26
    çľ¼çķĮ
    0.24
    éĻĦåĬł
    0.24
     görÃ¼ÅŁ
    0.24
     Front
    0.24
    arked
    0.24
     Downing
    0.23
    åĢŃ
    0.23
    ieÅĦ
    0.23
     "%
    0.23
    Activations Density 0.013%

    No Known Activations