© 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. 27-TRANSCODER-HP
    7. 8330
    Prev
    Next
    INDEX
    Explanations

    scientific equipment

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_27.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.27.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     СообÑī
    -0.28
    riminal
    -0.27
    .DataBind
    -0.26
    æİīèIJ½
    -0.26
    -UA
    -0.26
     usern
    -0.26
    責
    -0.26
    icode
    -0.25
     Dealer
    -0.25
    æ¶ĵ
    -0.25
    POSITIVE LOGITS
     satellites
    0.33
     machines
    0.31
     filled
    0.31
     towers
    0.30
     equipment
    0.29
    åį«æĺŁ
    0.29
     detectors
    0.29
     equipped
    0.28
    oscope
    0.28
     satellite
    0.28
    Activations Density 0.002%

    No Known Activations