© 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. 96214
    Prev
    Next
    INDEX
    Explanations

    teach

    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
    stery
    -0.28
    æĶ¶
    -0.27
    xes
    -0.27
    åįģåĽĽæĿ¡
    -0.27
    æºIJ头
    -0.26
    akis
    -0.26
    å¼Ģåı£
    -0.26
    åIJĪèµĦ
    -0.26
    èĽĽ
    -0.25
    åŃ¦ä¹ł
    -0.25
    POSITIVE LOGITS
     how
    0.45
     what
    0.37
    å¦Ĥä½ķ
    0.35
    how
    0.33
     kids
    0.32
     relaxation
    0.31
     social
    0.31
    æĢİä¹Ī
    0.29
    apa
    0.29
    åŃ©åŃIJ
    0.29
    Activations Density 0.103%

    No Known Activations