© 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. 35-TRANSCODER-HP
    7. 39665
    Prev
    Next
    INDEX
    Explanations

    code/technical text

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    é¦ĸåħĪè¦ģ
    -0.29
     recently
    -0.28
    åħĪ
    -0.28
    days
    -0.27
     Would
    -0.26
    åľ¨æѤ
    -0.26
     first
    -0.25
    cho
    -0.25
    Would
    -0.25
    ãģĵãĤĮãģĭãĤī
    -0.25
    POSITIVE LOGITS
     tablename
    0.29
    urban
    0.26
    çļĦåij½è¿IJ
    0.26
     propName
    0.25
     hashmap
    0.25
    æĹıèĩªæ²»åİ¿
    0.24
     loophole
    0.24
    çĶ²éĨĽ
    0.24
     evapor
    0.24
     oppressive
    0.24
    Activations Density 8.624%

    No Known Activations