© 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. 23-TRANSCODER-HP
    7. 25454
    Prev
    Next
    INDEX
    Explanations

    from

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_23.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.23.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     rew
    -0.30
    ard
    -0.28
     оÑĩеÑĢед
    -0.27
     IU
    -0.26
    indow
    -0.25
     control
    -0.25
    able
    -0.25
    ä¾ĭè¡Į
    -0.24
     alma
    -0.24
     Humb
    -0.24
    POSITIVE LOGITS
    ä¸Ģå¼Ģå§ĭå°±
    0.69
    ä¸Ģå¼Ģå§ĭ
    0.58
    å§ĭç»Ī
    0.41
    ä¸Ģ缴éĥ½
    0.40
    以æĿ¥
    0.39
    ä¸Ģ缴æĺ¯
    0.38
    ä¸Ģ缴éĥ½æĺ¯
    0.38
    ä¸Ģ缴
    0.38
     outset
    0.38
     dès
    0.37
    Activations Density 0.016%

    No Known Activations