© 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. 19320
    Prev
    Next
    INDEX
    Explanations

    Non-English characters/words

    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
    (ord
    -0.28
    æĶ¯
    -0.28
     root
    -0.26
     Impro
    -0.26
     inf
    -0.26
    imations
    -0.25
    ordin
    -0.25
    inf
    -0.25
     spir
    -0.25
     carbonate
    -0.25
    POSITIVE LOGITS
    çĽĺ
    0.30
    .mj
    0.29
    èĿł
    0.28
    ç©¿
    0.27
    çĭIJ
    0.27
    å¢ŀæĶ¶
    0.27
    è¢į
    0.26
    éĤĬ
    0.26
    èĪ«
    0.25
    fant
    0.25
    Activations Density 1.049%

    No Known Activations