© 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. 4716
    Prev
    Next
    INDEX
    Explanations

    modify

    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
    æ¤ħ
    -0.28
    aying
    -0.28
    çľĭä¸įè§ģ
    -0.27
    ruk
    -0.26
    lush
    -0.26
    ijk
    -0.26
     unintended
    -0.26
    ứng
    -0.26
     chess
    -0.25
    akk
    -0.25
    POSITIVE LOGITS
     Conf
    0.25
    åŁ¹
    0.24
     sor
    0.24
     hele
    0.24
    æĥ¯
    0.23
    åŃ°
    0.23
    sstream
    0.23
     numel
    0.23
     confer
    0.23
     pac
    0.23
    Activations Density 0.026%

    No Known Activations