© 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. 89327
    Prev
    Next
    INDEX
    Explanations

    code/technical documents

    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
    ahu
    -0.26
     Elem
    -0.26
    antha
    -0.25
    bery
    -0.25
    æĢĢæĬ±
    -0.25
    verture
    -0.25
    ystate
    -0.24
    atu
    -0.24
     pup
    -0.24
    TYPO
    -0.24
    POSITIVE LOGITS
     collections
    0.28
     mish
    0.26
     collection
    0.26
     crow
    0.26
    çļĦæłĩåĩĨ
    0.26
     partner
    0.25
     following
    0.24
    å¼¯æĽ²
    0.24
     Collections
    0.23
     direction
    0.23
    Activations Density 54.250%

    No Known Activations