© 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. 31-TRANSCODER-HP
    7. 18235
    Prev
    Next
    INDEX
    Explanations

    families and children

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Dirk
    -0.29
    STS
    -0.27
    çŁ³å®¶
    -0.26
    éŃĶ鬼
    -0.26
     Lisa
    -0.26
    ủng
    -0.25
    ä¸Ńè¶ħ
    -0.25
    ç§ijæĬĢèĤ¡ä»½
    -0.25
     <!--[
    -0.25
    becca
    -0.25
    POSITIVE LOGITS
    éĺħ
    0.37
    äºĪ
    0.34
    Asked
    0.32
    åĪĿå¿ĥ
    0.29
     Ev
    0.28
    éĸ±
    0.28
    åķı
    0.27
    unning
    0.27
    æ¸ħ
    0.27
    æĭ¾
    0.27
    Activations Density 4.813%

    No Known Activations