© 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. 96255
    Prev
    Next
    INDEX
    Explanations

    Clinical study comparisons

    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
     healthy
    -1.09
    åģ¥åº·çļĦ
    -1.02
    healthy
    -0.92
    æŃ£å¸¸
    -0.89
     Healthy
    -0.87
    Healthy
    -0.84
    æŃ£å¸¸çļĦ
    -0.84
    èī¯å¥½çļĦ
    -0.82
     healthier
    -0.81
    åģ¥åº·
    -0.81
    POSITIVE LOGITS
    åķ¬
    0.27
    hound
    0.25
    ä¹Łä¸įè¦ģ
    0.24
    æĸŁ
    0.23
    nze
    0.23
     Sie
    0.23
     UIBarButtonItem
    0.23
    èķ°
    0.23
    çļĦåľ°æŃ¥
    0.23
    é¦ĸåĪĽ
    0.22
    Activations Density 0.924%

    No Known Activations