© 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. 31-TRANSCODER-HP
    7. 17913
    Prev
    Next
    INDEX
    Explanations

    Lighting/candles

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     domina
    -0.30
     domination
    -0.30
     пÑĥ
    -0.29
     accomp
    -0.26
    çĥŃ
    -0.26
    éĿ¢
    -0.25
    è¿ĩ硬
    -0.25
    -fire
    -0.25
     firing
    -0.24
    ropolis
    -0.24
    POSITIVE LOGITS
     Trim
    0.28
    ãĥĻãĥ¼ãĤ¹
    0.28
     BaseEntity
    0.26
    trimmed
    0.26
     base
    0.26
    ç½ijåıĭè¯Ħ论
    0.26
    Trim
    0.25
    base
    0.25
    íĻĢ
    0.25
     bases
    0.25
    Activations Density 0.142%

    No Known Activations