© 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. 53821
    Prev
    Next
    INDEX
    Explanations

    ;

    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
     BA
    -0.41
    KA
    -0.34
     WA
    -0.34
     KA
    -0.33
     BAB
    -0.33
    BAB
    -0.33
     HA
    -0.32
    HA
    -0.32
     RA
    -0.32
     Mam
    -0.31
    POSITIVE LOGITS
    evity
    0.28
     ÑģÑĤаÑĤÑĮ
    0.28
    ierge
    0.26
    pillar
    0.25
    udios
    0.25
    elsius
    0.25
    ories
    0.25
     Pills
    0.25
    izzard
    0.24
    为导åIJij
    0.24
    Activations Density 0.274%

    No Known Activations