© 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. 107058
    Prev
    Next
    INDEX
    Explanations

    Code, languages, images, databases

    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
    onto
    -0.26
    Beer
    -0.26
    obel
    -0.26
    mar
    -0.26
    acf
    -0.25
    obl
    -0.25
     bic
    -0.25
    Filled
    -0.25
    formed
    -0.24
     glBind
    -0.24
    POSITIVE LOGITS
    æīĢæīĢ
    0.28
    illes
    0.26
    ú
    0.25
    tÃŃ
    0.25
    temporary
    0.25
    tips
    0.24
     existed
    0.24
    åij¤
    0.24
    æĹłæķ°æ¬¡
    0.23
    ingles
    0.23
    Activations Density 0.083%

    No Known Activations