© 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. 106919
    Prev
    Next
    INDEX
    Explanations

    conditional words

    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
    Negative Logits
     anything
    -0.31
    ä»»ä½ķ
    -0.29
    anything
    -0.29
    ä»»ä½ķ人
    -0.28
     anywhere
    -0.28
    ä¸įåĨįæĺ¯
    -0.27
    ä»Ģä¹Īäºĭ
    -0.27
    åĵªæĢķæĺ¯
    -0.26
    ä¸įå°ıçļĦ
    -0.26
    ä»Ģä¹Īåľ°æĸ¹
    -0.26
    POSITIVE LOGITS
     Hand
    0.28
     oranges
    0.26
    PEC
    0.26
    alyzed
    0.26
    pec
    0.26
    -relative
    0.25
    uccess
    0.25
     hand
    0.25
    çĶ¨æīĭ
    0.24
    salt
    0.24
    Activations Density 0.935%

    No Known Activations