© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 31-TRANSCODER-HP
    7. 17748
    Prev
    Next
    INDEX
    Explanations

    Non-English tokens

    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
    Negative Logits
    lass
    -0.28
    ıl
    -0.28
     trait
    -0.27
     isolation
    -0.26
    çłĶ
    -0.26
    ä¸įè¿ĩ
    -0.26
    lover
    -0.26
    éħ¬
    -0.26
    cue
    -0.26
     unst
    -0.26
    POSITIVE LOGITS
    insky
    0.28
    >]
    0.27
    èĴ²
    0.26
    èľ¥
    0.25
    åĩºçīĪ
    0.25
    %C
    0.24
    }]
    0.24
    ieres
    0.24
    æīĩ
    0.24
    ä¸Ńæ¯Ĵ
    0.23
    Activations Density 0.061%

    No Known Activations