© 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. 95446
    Prev
    Next
    INDEX
    Explanations

    well

    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
     peoples
    -0.47
     Groups
    -0.36
     People
    -0.35
     groups
    -0.35
     people
    -0.34
     Members
    -0.34
    People
    -0.33
    groups
    -0.33
    people
    -0.32
     Humans
    -0.32
    POSITIVE LOGITS
    类产åĵģ
    0.28
    åIJĪä¸Ģ
    0.27
    owie
    0.27
    bee
    0.27
     Queen
    0.27
     dock
    0.26
    çĶµ
    0.26
     frontier
    0.25
    queen
    0.25
     queen
    0.25
    Activations Density 0.893%

    No Known Activations