© 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. 96423
    Prev
    Next
    INDEX
    Explanations

    4

    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
     card
    -0.26
    esch
    -0.25
     bridge
    -0.25
    亲
    -0.25
    card
    -0.25
     rig
    -0.25
    dro
    -0.24
     discard
    -0.24
     arc
    -0.24
    CARD
    -0.24
    POSITIVE LOGITS
    ä½ĵ
    0.29
    itten
    0.28
    åĩ½
    0.27
    ampoline
    0.26
    odium
    0.26
    monic
    0.25
    formance
    0.25
    paragus
    0.25
    odom
    0.24
    åĢĴ
    0.24
    Activations Density 0.907%

    No Known Activations