© 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. 109261
    Prev
    Next
    INDEX
    Explanations

    Y

    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
    Not in Any Lists

    No Comments

    Negative Logits
    ä¸ĭè¡Į
    -0.30
    cestor
    -0.26
    å·±
    -0.25
    ernet
    -0.25
    reserved
    -0.25
    æĥħåķĨ
    -0.25
    EXTERN
    -0.25
    EDIT
    -0.25
     selves
    -0.25
    (Box
    -0.25
    POSITIVE LOGITS
     strchr
    0.27
    leck
    0.25
    ÅĤu
    0.25
     towns
    0.24
     shallow
    0.24
    ::*
    0.24
    -mini
    0.23
    è··
    0.23
    å½¹
    0.23
     afford
    0.23
    Activations Density 0.340%

    No Known Activations