© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 33-TRANSCODER-HP
    7. 53618
    Prev
    Next
    INDEX
    Explanations

    technical articles

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ıģ
    -0.26
     stag
    -0.26
    _operator
    -0.26
    Ms
    -0.26
    aret
    -0.26
    online
    -0.26
     online
    -0.25
    ä¼łåĩº
    -0.25
     operator
    -0.25
    åŀĽ
    -0.24
    POSITIVE LOGITS
    RIGHT
    0.29
    imen
    0.29
    å·²ç»ı被
    0.28
    ylene
    0.26
    亵
    0.26
    gado
    0.26
     RIGHT
    0.26
    dio
    0.25
     tubing
    0.25
     inter
    0.25
    Activations Density 0.007%

    No Known Activations