© 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. 25-TRANSCODER-HP
    7. 107044
    Prev
    Next
    INDEX
    Explanations

    excel

    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
    Negative Logits
     пÑĢоведен
    -0.28
    done
    -0.26
    伤
    -0.26
     Marks
    -0.25
     agreement
    -0.25
     agre
    -0.24
    ni
    -0.24
    .done
    -0.24
     Fel
    -0.24
    qui
    -0.24
    POSITIVE LOGITS
    äºĨ好
    0.30
    /vnd
    0.25
    æĹ¥äº§
    0.25
     dop
    0.24
    楯
    0.24
    Ä°Z
    0.24
    tog
    0.23
    OUNDS
    0.23
    KD
    0.23
    //-----------------------------------------------------------------------------↵
    0.23
    Activations Density 1.012%

    No Known Activations