© 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. 95771
    Prev
    Next
    INDEX
    Explanations

    Thomas

    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
    山寨
    -0.32
     convergence
    -0.27
     agree
    -0.27
    éĵ¶
    -0.25
    ibold
    -0.25
    åĨľåī¯
    -0.25
     agreed
    -0.25
    à¹ģห
    -0.25
    æŁ¥
    -0.25
     Black
    -0.24
    POSITIVE LOGITS
     Buchanan
    0.29
     pot
    0.28
    enders
    0.28
    åĬĪ
    0.26
    _AES
    0.26
    ICES
    0.26
    olley
    0.25
    å»īæĶ¿
    0.24
     lawmaker
    0.24
     SWAT
    0.24
    Activations Density 0.005%

    No Known Activations