© 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. 30-TRANSCODER-HP
    7. 95020
    Prev
    Next
    INDEX
    Explanations

    adjust

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    款
    -0.27
     Libert
    -0.27
    reflect
    -0.26
    æĸ½
    -0.25
     Vict
    -0.25
     flats
    -0.25
    pins
    -0.24
    authorize
    -0.24
    åı¥
    -0.24
     reflect
    -0.23
    POSITIVE LOGITS
     Adjustment
    0.40
     adjustment
    0.39
    è°ĥæķ´
    0.36
     adjusting
    0.35
     adjust
    0.33
     adjustments
    0.32
    adjust
    0.31
     ajust
    0.31
    Adjust
    0.30
    ç¼ĸè¾ij
    0.29
    Activations Density 0.887%

    No Known Activations