© 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. 107194
    Prev
    Next
    INDEX
    Explanations

    code and unusual text

    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
     comp
    -0.30
    ats
    -0.30
    å²Ľ
    -0.27
     expect
    -0.26
     steps
    -0.26
     expectation
    -0.25
     sharing
    -0.25
    ãģľ
    -0.25
     Steps
    -0.25
    Steps
    -0.25
    POSITIVE LOGITS
    icer
    0.25
    CAPE
    0.25
    iod
    0.25
    è¯ģ书
    0.24
    ounsel
    0.24
    dae
    0.24
    æľīå¤ļ大
    0.24
    @(
    0.24
    ued
    0.24
     Certificate
    0.24
    Activations Density 0.151%

    No Known Activations