© 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. 35-TRANSCODER-HP
    7. 38705
    Prev
    Next
    INDEX
    Explanations

    -de suffix

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     move
    -0.29
     change
    -0.28
     overs
    -0.26
     overhead
    -0.26
    çĪIJ
    -0.25
     theoretically
    -0.25
     differently
    -0.25
     upwards
    -0.25
     enormous
    -0.25
    æĥ³è¦ģ
    -0.25
    POSITIVE LOGITS
    central
    0.33
    cross
    0.31
    care
    0.31
    caps
    0.30
    composition
    0.30
    cod
    0.30
    composed
    0.30
    secondary
    0.30
    cup
    0.30
    corner
    0.29
    Activations Density 10.048%

    No Known Activations