© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 25-TRANSCODER-HP
    7. 107221
    Prev
    Next
    INDEX
    Explanations

    Online excerpts

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     oneself
    -0.69
    æĬĬä½ł
    -0.67
    ä½łçļĦ
    -0.65
     yourself
    -0.63
     your
    -0.61
    your
    -0.59
    ä½ł
    -0.59
    ç»Ļä½ł
    -0.56
    è®©ä½ł
    -0.56
    ä¸ºä½ł
    -0.55
    POSITIVE LOGITS
    .Future
    0.26
    Xml
    0.25
    ahead
    0.24
    åıijçĶŁäºĨ
    0.24
    dent
    0.24
    éħ¬
    0.23
    对该
    0.23
     ký
    0.23
    losures
    0.23
    åĻĻ
    0.23
    Activations Density 0.907%

    No Known Activations