© 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. 95860
    Prev
    Next
    INDEX
    Explanations

    Changes in position

    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
    æĴ¤
    -0.94
    åį¸
    -0.92
    æĴ¤éĶĢ
    -0.88
     removal
    -0.87
    æĭĨéĻ¤
    -0.84
     removing
    -0.81
    解éĻ¤
    -0.80
    åıĸæ¶Ī
    -0.79
     remove
    -0.77
     withdrawal
    -0.75
    POSITIVE LOGITS
    æĿµ
    0.28
    ç¬Ļ
    0.24
     improvised
    0.24
    åŃ³
    0.23
    elle
    0.22
    å¢ŀ强
    0.22
    addin
    0.22
    èĢģ人家
    0.22
     bisc
    0.21
    anding
    0.21
    Activations Density 0.913%

    No Known Activations