© 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. 24833
    Prev
    Next
    INDEX
    Explanations

    references

    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
    itat
    -0.30
     fact
    -0.28
    éĢĹ
    -0.28
    utta
    -0.24
    æĶ¶
    -0.24
    -fit
    -0.24
    >(()
    -0.24
    ization
    -0.24
    èĭ¥æĺ¯
    -0.24
    .mit
    -0.24
    POSITIVE LOGITS
     OMIT
    0.29
    éĥ¨
    0.26
     Calendar
    0.26
    è¿«
    0.26
     alphabetical
    0.25
    expense
    0.25
     ìĤ¬ìĹħ
    0.25
     Dü
    0.24
    essed
    0.24
     kino
    0.23
    Activations Density 0.069%

    No Known Activations