© 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. 30-TRANSCODER-HP
    7. 96205
    Prev
    Next
    INDEX
    Explanations

    code and terminal commands

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rod
    -0.28
    åĽĽçº§
    -0.26
    \Migration
    -0.25
    åĵĪ
    -0.25
    软
    -0.25
    CHED
    -0.25
    æ´¥
    -0.25
    姿åĬ¿
    -0.24
     hashCode
    -0.24
    Opera
    -0.23
    POSITIVE LOGITS
     ./
    0.33
    ials
    0.31
    ./
    0.28
     bundle
    0.28
    als
    0.27
    éĩı
    0.27
     ev
    0.26
     python
    0.26
     Rah
    0.26
    imum
    0.25
    Activations Density 0.986%

    No Known Activations