© 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. 53672
    Prev
    Next
    INDEX
    Explanations

    places

    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
     Feinstein
    -0.29
    моÑĤ
    -0.29
    eca
    -0.28
    Gatt
    -0.28
    GIT
    -0.27
     spot
    -0.26
    Kid
    -0.26
    eor
    -0.25
    IFF
    -0.25
    ifo
    -0.25
    POSITIVE LOGITS
    çĻ¾ä¸ĩ
    0.29
    BS
    0.26
    ule
    0.26
    uttle
    0.24
    ness
    0.24
     Bs
    0.24
     British
    0.24
    assy
    0.24
    礴
    0.23
     buz
    0.23
    Activations Density 0.037%

    No Known Activations