© 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. 25-TRANSCODER-HP
    7. 107999
    Prev
    Next
    INDEX
    Explanations

    foreign words

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    åįļ士
    -0.29
     dissert
    -0.27
    edium
    -0.27
     very
    -0.27
    visa
    -0.26
     equiv
    -0.26
    åīįåı°
    -0.25
    èĪĨ
    -0.24
     Caul
    -0.24
     editorial
    -0.24
    POSITIVE LOGITS
    bedo
    0.29
    âľį
    0.28
    åĽĽå¤§
    0.28
    æĪijä¼ļ
    0.25
    endar
    0.25
    çī¹æĸ¯
    0.24
    主ä¹īèĢħ
    0.24
    æĬĵ
    0.24
    主ä¹ī
    0.24
    æĢ§æł¼
    0.24
    Activations Density 0.898%

    No Known Activations