© 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. 25-TRANSCODER-HP
    7. 108790
    Prev
    Next
    INDEX
    Explanations

    Academic/Descriptive texts

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     cref
    -0.26
    åij¨æľŁ
    -0.25
    ])-
    -0.24
    )))),
    -0.24
    lö
    -0.24
    yk
    -0.24
     WK
    -0.24
    tsy
    -0.23
     ck
    -0.23
     Mp
    -0.23
    POSITIVE LOGITS
    ishes
    0.30
    ä½ĨçĶ±äºİ
    0.29
    éĹªéĹª
    0.28
    è¿Ļä¸ĢçĤ¹
    0.28
    ä½Ĩåľ¨
    0.28
    itarian
    0.27
    лÑıеÑĤ
    0.27
    ÅŁe
    0.27
    zee
    0.26
    ä½Ĩæĺ¯åľ¨
    0.26
    Activations Density 0.928%

    No Known Activations