© 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. 52975
    Prev
    Next
    INDEX
    Explanations

    future

    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
    vana
    -0.32
    atra
    -0.31
     :";↵
    -0.29
    romo
    -0.29
     nextPage
    -0.27
    èĸħ
    -0.27
    GridColumn
    -0.26
    尾巴
    -0.26
    å±ŀäºİèĩªå·±
    -0.26
    evice
    -0.26
    POSITIVE LOGITS
     effects
    0.31
     planning
    0.31
     age
    0.30
     talents
    0.30
     strikes
    0.30
    --
    0.30
     interactions
    0.30
     growth
    0.29
     relationships
    0.29
     courses
    0.29
    Activations Density 0.012%

    No Known Activations