© 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. 35-TRANSCODER-HP
    7. 40420
    Prev
    Next
    INDEX
    Explanations

    Code and technical lists

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    opc
    -0.31
    æīįæĺ¯
    -0.28
     CONSEQUENTIAL
    -0.27
    esion
    -0.26
    ecer
    -0.26
    ä¸Ģ个æĺ¯
    -0.26
    ische
    -0.26
    ocha
    -0.25
    æĺ¯æľī
    -0.25
    ischer
    -0.25
    POSITIVE LOGITS
     Solution
    0.29
     Reve
    0.28
     Shoulder
    0.27
     replied
    0.27
     conform
    0.26
     Modified
    0.26
     Ste
    0.26
     Match
    0.26
     loops
    0.26
     attenu
    0.25
    Activations Density 0.062%

    No Known Activations