© 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. 31-TRANSCODER-HP
    7. 18197
    Prev
    Next
    INDEX
    Explanations

    communication systems

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    象
    -0.28
    åı°
    -0.28
    罪
    -0.27
    æī¬
    -0.26
    -messages
    -0.25
    äºĴ
    -0.24
    piler
    -0.24
    redentials
    -0.24
     ins
    -0.24
    orch
    -0.24
    POSITIVE LOGITS
     APS
    0.35
     diversity
    0.32
     Attr
    0.30
    å¤ļæł·æĢ§
    0.27
    arium
    0.27
     MPS
    0.27
    æĴ®
    0.27
    APS
    0.26
    åĩĭ
    0.26
    å¤ļæł·åĮĸ
    0.26
    Activations Density 1.714%

    No Known Activations