© 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. 107203
    Prev
    Next
    INDEX
    Explanations

    what

    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
    使
    -0.28
     SUBJECT
    -0.28
    ä½ľä¸º
    -0.27
    åıĹ
    -0.27
    说
    -0.27
    æł¡
    -0.26
    .getBoolean
    -0.26
    å½ĵçĦ¶
    -0.25
    ä½ľçĤº
    -0.25
    梯
    -0.25
    POSITIVE LOGITS
    ure
    0.27
    urf
    0.27
    idf
    0.26
    ifest
    0.24
    çļĦå¿ĥæĢģ
    0.24
    ç»Ļä½łä»¬
    0.24
     accountable
    0.23
    UTES
    0.23
    ssf
    0.23
    oru
    0.23
    Activations Density 0.969%

    No Known Activations