© 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. 12-TRANSCODER-HP
    7. 22960
    Prev
    Next
    INDEX
    Explanations

    android

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_12.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.12.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ç¦ģåĮº
    -0.31
    ç¿ķ
    -0.27
    Nb
    -0.26
    åħ¬è®¤çļĦ
    -0.26
    çĵ¶
    -0.25
    åĴ«
    -0.25
    åĸľå¥½
    -0.25
    èĥĮçĿĢ
    -0.24
    è¸Į
    -0.24
     needle
    -0.24
    POSITIVE LOGITS
     Blo
    0.27
     release
    0.27
    è¿ĩåī©
    0.26
    retch
    0.25
    çī¯
    0.25
    ley
    0.25
    keley
    0.25
    ');</
    0.25
    oute
    0.25
    )</
    0.25
    Activations Density 0.002%

    No Known Activations