© 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. 98627
    Prev
    Next
    INDEX
    Explanations

    hospital

    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
     often
    -0.45
    often
    -0.40
     usually
    -0.39
     Often
    -0.39
    å¾Ģå¾Ģ
    -0.36
     souvent
    -0.34
     Usually
    -0.34
     typically
    -0.32
     seldom
    -0.32
    Often
    -0.32
    POSITIVE LOGITS
     possibly
    0.56
     potentially
    0.49
    æľīåı¯èĥ½
    0.47
     could
    0.44
    åı¯èĥ½ä¼ļ
    0.43
    possibly
    0.42
     Possibly
    0.42
    åı¯èĥ½
    0.41
    åı¯èĥ½æĢ§
    0.40
     perhaps
    0.39
    Activations Density 0.002%

    No Known Activations