© 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. 20-TRANSCODER-HP
    7. 142124
    Prev
    Next
    INDEX
    Explanations

    role

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    inite
    -0.27
    æ°ª
    -0.26
    enses
    -0.26
     accommod
    -0.26
    æļĤæŶ
    -0.25
    formats
    -0.25
     Mane
    -0.24
    è¿ĺ好
    -0.24
    æĦŁ
    -0.24
    è¦ģçĤ¹
    -0.24
    POSITIVE LOGITS
    eer
    0.29
    r
    0.28
     cross
    0.27
     crosses
    0.27
    .getBody
    0.27
    ons
    0.25
     overlapping
    0.24
    /body
    0.24
    oven
    0.24
    pga
    0.24
    Activations Density 6.841%

    No Known Activations