© 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. 106981
    Prev
    Next
    INDEX
    Explanations

    iHeartMedia

    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
    æıī
    -0.29
    rena
    -0.27
    æ³·
    -0.27
     Mapper
    -0.27
    åĽłåľ°
    -0.26
     Hobby
    -0.26
    ProcessEvent
    -0.26
    ç»ıæŁ¥
    -0.24
    æĪij家
    -0.24
    ä¸Ģ个éĹ®é¢ĺ
    -0.24
    POSITIVE LOGITS
    azine
    0.31
    hip
    0.27
    åł¡
    0.25
    lig
    0.23
     com
    0.23
    ifes
    0.23
     recogn
    0.23
    ics
    0.23
     couple
    0.22
     UN
    0.22
    Activations Density 0.996%

    No Known Activations