© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 19-TRANSCODER-HP
    7. 7006
    Prev
    Next
    INDEX
    Explanations

    Code, paths, names, debug messages

    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
     strain
    -0.26
    oven
    -0.25
    ifter
    -0.25
    grounds
    -0.24
    otion
    -0.24
    imeo
    -0.23
    .Complete
    -0.23
    æ³¢çī¹
    -0.23
    åıªèĥ½
    -0.23
    -only
    -0.23
    POSITIVE LOGITS
    åĢį
    0.25
    å±Ĥåĩº
    0.24
     Ãģ
    0.24
    çŁ³å®¶
    0.23
     Description
    0.23
    ![
    0.23
    ë¡
    0.23
    çĭIJ
    0.23
     TER
    0.23
    Ñģло
    0.23
    Activations Density 0.059%

    No Known Activations