© 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. 22-TRANSCODER-HP
    7. 130097
    Prev
    Next
    INDEX
    Explanations

    Word fragments

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.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·¤
    -0.26
    missing
    -0.25
     herself
    -0.23
    åı±
    -0.23
    Servi
    -0.23
     playable
    -0.23
     bleed
    -0.23
    åĩłä¹İæĺ¯
    -0.23
     implanted
    -0.22
     beds
    -0.22
    POSITIVE LOGITS
    aise
    0.29
    eries
    0.26
    STA
    0.26
    ç¹ģ
    0.25
    貫
    0.24
    uns
    0.24
     pul
    0.24
     Davies
    0.24
    rollo
    0.24
     laps
    0.24
    Activations Density 0.089%

    No Known Activations