© 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. 33-TRANSCODER-HP
    7. 54436
    Prev
    Next
    INDEX
    Explanations

    specify location or features

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    éĩİçĶŁ
    -0.26
    igram
    -0.25
    cis
    -0.25
    ä¸Ģæ³¢
    -0.25
    èĩªæĦ¿
    -0.24
    éŁ³
    -0.23
    ssel
    -0.23
    lover
    -0.23
    amedi
    -0.23
    SuccessListener
    -0.23
    POSITIVE LOGITS
     detail
    0.31
     gro
    0.29
     positions
    0.27
    etu
    0.27
     details
    0.27
     locations
    0.27
     distinguishing
    0.27
     minimized
    0.26
     properties
    0.26
    min
    0.25
    Activations Density 0.018%

    No Known Activations