© 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. 95556
    Prev
    Next
    INDEX
    Explanations

    Red

    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
    æħ¢
    -0.43
     slowing
    -0.43
     slow
    -0.40
    Slow
    -0.39
     slows
    -0.39
    slow
    -0.39
     Slow
    -0.36
     slowed
    -0.32
     slower
    -0.32
    åģľå·¥
    -0.31
    POSITIVE LOGITS
     Red
    0.31
     moment
    0.28
    red
    0.28
    ComputedStyle
    0.27
     relatively
    0.27
     relative
    0.27
    naire
    0.27
    _req
    0.27
    Red
    0.26
     red
    0.26
    Activations Density 0.911%

    No Known Activations