© 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. 111408
    Prev
    Next
    INDEX
    Explanations

    long, formal documents

    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
    ä¸įéľĢè¦ģ
    -0.28
    æłĩåĩĨ
    -0.28
    èĢĮä¸į
    -0.27
    ä¸į大
    -0.26
    ä¸į太好
    -0.25
    ä¸įçĶ¨
    -0.25
    æĪij没æľī
    -0.25
    ä¸į容æĺĵ
    -0.25
    çĭ¬
    -0.25
    并éĢļè¿ĩ
    -0.24
    POSITIVE LOGITS
    cream
    0.24
    зв
    0.24
    Rich
    0.24
    Procedure
    0.24
    fty
    0.24
    ife
    0.24
    Numer
    0.24
    oze
    0.24
    rer
    0.24
     ASAP
    0.23
    Activations Density 0.005%

    No Known Activations