© 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. 23-TRANSCODER-HP
    7. 24819
    Prev
    Next
    INDEX
    Explanations

    Past-tense verbs

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_23.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.23.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    iban
    -0.28
    æĢ§çĸ¾çĹħ
    -0.27
    oteric
    -0.27
    æĢ§çļĦ
    -0.26
    è¯Ģ
    -0.26
    itives
    -0.26
    æĢ§åĴĮ
    -0.25
    åŀĭä¼ģä¸ļ
    -0.25
    ä¸Ģåľº
    -0.24
    æľīæĦıä¹ī
    -0.24
    POSITIVE LOGITS
     evidence
    0.36
     evidenced
    0.36
    Evidence
    0.26
     evid
    0.25
    ä¿¡
    0.25
    .@
    0.25
     cuent
    0.25
    æĬĹè®®
    0.25
     handling
    0.24
     Witness
    0.24
    Activations Density 0.006%

    No Known Activations