© 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. 32-TRANSCODER-HP
    7. 43293
    Prev
    Next
    INDEX
    Explanations

    carbohydrate

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    arro
    -0.29
    ighter
    -0.27
    irit
    -0.26
    umbotron
    -0.25
    ++++++++++++++++++++++++++++++++
    -0.25
    ´t
    -0.25
    anter
    -0.24
    endi
    -0.24
    iced
    -0.24
     trous
    -0.24
    POSITIVE LOGITS
     complex
    0.42
     Complex
    0.38
    complex
    0.36
    è¤ĩ
    0.36
    Complex
    0.34
     free
    0.31
    å¤į
    0.31
    å¤įæĿĤ
    0.31
    å¤įæĿĤçļĦ
    0.30
    å¤įåIJĪ
    0.30
    Activations Density 0.003%

    No Known Activations