© 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. 24-TRANSCODER-HP
    7. 151111
    Prev
    Next
    INDEX
    Explanations

    informal reflections

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_24.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.24.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ä¹ĭ举
    -0.30
    ise
    -0.30
    è¿ĻäºĽéĹ®é¢ĺ
    -0.29
    CUS
    -0.28
     humble
    -0.28
    ITS
    -0.28
     sá»±
    -0.26
    utive
    -0.25
    é«ĺç´łè´¨
    -0.25
    vestment
    -0.25
    POSITIVE LOGITS
     having
    0.62
     being
    0.60
     making
    0.59
     needing
    0.51
    having
    0.49
     getting
    0.48
     keeping
    0.48
     doing
    0.47
     ensuring
    0.46
     taking
    0.45
    Activations Density 1.381%

    No Known Activations