© 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. 89802
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    çĶļä¹Ī
    -0.26
    æŃ£å®Ĺ
    -0.25
     Zot
    -0.25
    Nib
    -0.25
    ä¸Ģæľ¬ä¹¦
    -0.24
    èĢģ头
    -0.24
    å®īåħ¨ä¿Ŀéļľ
    -0.23
    çļĦä½ľåĵģ
    -0.23
    latlong
    -0.23
    ä¸įéĹ´
    -0.23
    POSITIVE LOGITS
     ha
    0.26
     Keystone
    0.25
     pitch
    0.24
     measure
    0.24
     Measurement
    0.24
    åĤĻ
    0.24
     Shift
    0.23
     shift
    0.23
    åIJł
    0.23
     Standard
    0.23
    Activations Density 0.025%

    No Known Activations

    This feature has no known activations.