© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 18-TRANSCODER-HP
    7. 89712
    Prev
    Next
    INDEX
    Explanations

    communities coming together

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Selective
    -0.28
    空æ°Ķ
    -0.25
    è®°å¿Ĩ
    -0.25
    èĩĮ
    -0.24
     Olympia
    -0.24
    æݪ
    -0.24
    ILog
    -0.24
     autos
    -0.24
    éĢŁåº¦
    -0.24
    gatsby
    -0.23
    POSITIVE LOGITS
    ä¸Ģ身
    0.29
     intro
    0.26
    pole
    0.26
    åįĥåı¤
    0.26
    -door
    0.26
    ipur
    0.25
    -center
    0.24
    ious
    0.24
    éŨ
    0.24
     ^.
    0.24
    Activations Density 11.159%

    No Known Activations