© 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. 95334
    Prev
    Next
    INDEX
    Explanations

    works

    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
    çģ¸
    -0.26
     laying
    -0.25
    æĶ¶åĽŀ
    -0.25
    楼主
    -0.24
    åIJĥä»Ģä¹Ī
    -0.24
    quals
    -0.24
     storyline
    -0.24
    avel
    -0.24
    ForObject
    -0.24
    .put
    -0.24
    POSITIVE LOGITS
    å·¥ä½ľ
    0.34
    éķ¿
    0.34
    å·¥ä½ľçļĦ
    0.34
     long
    0.33
     work
    0.32
    å¿«
    0.31
    <long
    0.31
     according
    0.31
     smoothly
    0.30
    long
    0.30
    Activations Density 0.111%

    No Known Activations