© 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. 19-TRANSCODER-HP
    7. 8404
    Prev
    Next
    INDEX
    Explanations

    Research studies

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    walking
    -0.28
    饮çĶ¨æ°´
    -0.28
    樹
    -0.26
    orer
    -0.25
    å¾Īæ¸ħæ¥ļ
    -0.25
    带çĿĢ
    -0.25
    thur
    -0.24
     Shade
    -0.24
     sticks
    -0.24
    æİĪ
    -0.24
    POSITIVE LOGITS
    åīįæīĢ
    0.30
    -parts
    0.29
    åIJįçīĮ
    0.25
    :Register
    0.25
    ToMany
    0.25
    éĥ¨åĪĨ
    0.25
     brethren
    0.25
    lier
    0.25
    unfinished
    0.24
     Himself
    0.24
    Activations Density 0.035%

    No Known Activations