© 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. 18724
    Prev
    Next
    INDEX
    Explanations

    past tense verbs

    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
    ä¸įéĶĻçļĦ
    -0.31
    åIJĪéĢĤçļĦ
    -0.30
    å¾Īæ£Ĵ
    -0.28
    å¾Īæ¸ħæ¥ļ
    -0.28
    活泼
    -0.28
    ä¸Ģç»ı
    -0.28
     suitable
    -0.27
    æ³ĵ
    -0.27
    åIJĪéĢĤ
    -0.26
    å¯ĨåĪĩ
    -0.25
    POSITIVE LOGITS
     leaves
    0.33
    leave
    0.29
     Leaves
    0.29
     leaving
    0.29
    stead
    0.28
    thood
    0.27
    å¿«
    0.27
    大å¦Ī
    0.26
     leave
    0.26
     statusBar
    0.25
    Activations Density 0.945%

    No Known Activations