© 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. 96111
    Prev
    Next
    INDEX
    Explanations

    code snippets

    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
     integrating
    -0.50
     optimizing
    -0.50
     managing
    -0.50
     evaluating
    -0.48
     navigating
    -0.46
     calculating
    -0.46
     researching
    -0.45
     accessing
    -0.45
     implementing
    -0.44
     constructing
    -0.44
    POSITIVE LOGITS
     beh
    0.30
     selection
    0.28
    ä¾Ľå¤§å®¶
    0.26
    磾
    0.25
    selection
    0.25
    éĺħ读åħ¨æĸĩ
    0.25
    èij«
    0.24
    eres
    0.24
    è¡ĮåĬ¨
    0.24
    æĢ¿
    0.24
    Activations Density 0.893%

    No Known Activations