© 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. 18-TRANSCODER-HP
    7. 88481
    Prev
    Next
    INDEX
    Explanations

    Code/technical snippets

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    onium
    -0.25
    ÅŁÄ±
    -0.23
     escap
    -0.22
    isch
    -0.22
     }}}
    -0.22
    主人
    -0.22
     Commons
    -0.21
    ');?>↵
    -0.21
     âĢ¦↵↵
    -0.21
    urable
    -0.21
    POSITIVE LOGITS
    Implemented
    0.23
     hÃłi
    0.23
    arius
    0.23
     Measurement
    0.23
     Mi
    0.22
    át
    0.21
    measurement
    0.21
     measurement
    0.21
    patch
    0.21
     Mist
    0.21
    Activations Density 94.668%

    No Known Activations