© 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. 94938
    Prev
    Next
    INDEX
    Explanations

    location and device services

    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
    encoder
    -0.28
    stad
    -0.27
    Decoder
    -0.27
    pv
    -0.26
    iry
    -0.26
    itat
    -0.26
    .WriteByte
    -0.26
     Isl
    -0.25
    ','%
    -0.25
     CreateUser
    -0.25
    POSITIVE LOGITS
    æĶ»åĬ¿
    0.29
     dyn
    0.28
    Trees
    0.27
    æħĪåĸĦ
    0.26
    chi
    0.25
    chk
    0.25
     trees
    0.24
    èĦ¸é¢Ĭ
    0.24
    æĺŁçº§éħĴåºĹ
    0.24
    )n
    0.24
    Activations Density 0.906%

    No Known Activations