© 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. 109415
    Prev
    Next
    INDEX
    Explanations

    resources

    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
     Permission
    -0.27
    åijĬè¯ī
    -0.26
    altar
    -0.25
    à·Ħ
    -0.25
    沧
    -0.25
    许åı¯
    -0.25
    alto
    -0.25
    è¿Ļ对
    -0.24
     again
    -0.24
    è¶ĭ
    -0.24
    POSITIVE LOGITS
     resources
    0.32
    .relative
    0.29
    èµĦæºIJ
    0.28
    磾
    0.28
    supports
    0.27
     getResource
    0.27
    protocols
    0.27
    resources
    0.26
     campaigns
    0.26
     resource
    0.25
    Activations Density 0.935%

    No Known Activations