© 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. 20-TRANSCODER-HP
    7. 140978
    Prev
    Next
    INDEX
    Explanations

    answer

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    اث
    -0.32
    ACK
    -0.28
    è§Ĥæij©
    -0.26
    ç¾ŀ
    -0.26
    falls
    -0.26
    ä¸įæķ¢
    -0.26
    å¦Ĥæŀľæ²¡æľī
    -0.25
    éĶĻè¿ĩ
    -0.25
    å¼¹
    -0.25
    è¾ľè´Ł
    -0.24
    POSITIVE LOGITS
    clerosis
    0.27
     THAN
    0.26
     than
    0.26
     somehow
    0.26
    dress
    0.26
    urgeon
    0.26
    yssey
    0.26
    æ²Ķ
    0.26
    kre
    0.25
    æĽ¿ãģĪ
    0.24
    Activations Density 16.442%

    No Known Activations