© 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. 19767
    Prev
    Next
    INDEX
    Explanations

    computer code

    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
    颡
    -0.29
    ãĥ£
    -0.26
    豪
    -0.26
    anka
    -0.25
     ours
    -0.25
     Sounds
    -0.24
    APE
    -0.24
    ÂŃt
    -0.24
    羣å¿ĥ
    -0.24
     pris
    -0.24
    POSITIVE LOGITS
    åĪĽéĢłåĬĽ
    0.27
    翳
    0.26
    olecule
    0.26
    esion
    0.24
     Sur
    0.24
     dem
    0.24
    setMessage
    0.24
    å´¦
    0.24
    andalone
    0.24
    arsed
    0.23
    Activations Density 0.002%

    No Known Activations