© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 18-TRANSCODER-HP
    7. 89267
    Prev
    Next
    INDEX
    Explanations

    }

    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
     Charges
    -0.24
     bordel
    -0.23
     Activation
    -0.23
     Reject
    -0.23
     Fucking
    -0.22
     libertin
    -0.22
     Positions
    -0.22
     Conspiracy
    -0.22
     dames
    -0.22
     sodom
    -0.21
    POSITIVE LOGITS
     dis
    0.27
     out
    0.27
     cross
    0.27
     immediately
    0.26
     rel
    0.25
     present
    0.25
     long
    0.25
     count
    0.25
     like
    0.25
     pre
    0.25
    Activations Density 33.482%

    No Known Activations