© 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. 53579
    Prev
    Next
    INDEX
    Explanations

    al

    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
    (qu
    -0.31
     qu
    -0.30
    qu
    -0.30
    -qu
    -0.28
    å°Ĭæķ¬
    -0.27
    ipi
    -0.27
     pu
    -0.26
    Å·
    -0.26
    quer
    -0.26
    cu
    -0.26
    POSITIVE LOGITS
    firm
    0.32
    -Clause
    0.29
    bia
    0.28
     firm
    0.26
    bed
    0.26
     Firm
    0.25
    ফ
    0.25
    atching
    0.25
     Dise
    0.25
    åĨįåİ»
    0.24
    Activations Density 0.088%

    No Known Activations