© 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. 107149
    Prev
    Next
    INDEX
    Explanations

    number

    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
    æĸŃ
    -0.31
    SEA
    -0.28
    åı£
    -0.26
    å±ĤéĿ¢
    -0.26
    æĸŃè£Ĥ
    -0.26
    ç©·
    -0.25
    çĶŁåij½çļĦ
    -0.25
    éŀĺ
    -0.25
     skype
    -0.25
    åı¥
    -0.25
    POSITIVE LOGITS
    agan
    0.27
    bru
    0.27
    hal
    0.27
     welded
    0.26
     halo
    0.26
    ertype
    0.26
    ference
    0.25
     corros
    0.25
    FUL
    0.25
    -inline
    0.25
    Activations Density 0.895%

    No Known Activations