© 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. 88549
    Prev
    Next
    INDEX
    Explanations

    Scientific/technical writing

    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
    -static
    -0.31
    çĴĭ
    -0.26
     Licence
    -0.26
     Goldberg
    -0.25
     DM
    -0.25
    çıī
    -0.25
     ach
    -0.24
     cheated
    -0.23
     crushing
    -0.23
    -breaking
    -0.23
    POSITIVE LOGITS
    æľīæīĢ
    0.27
    ULO
    0.25
    enable
    0.25
    ulant
    0.24
     cour
    0.24
    èĪŁ
    0.24
    ,np
    0.24
    ely
    0.24
    çĽĸ
    0.24
    Signals
    0.24
    Activations Density 80.176%

    No Known Activations