© 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. 22-TRANSCODER-HP
    7. 129916
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Fcn
    -0.28
    好å¤ļ
    -0.26
    auc
    -0.26
    ãĤŃãĥ£ãĥ³
    -0.26
    ãĥijãĥ¯ãĥ¼
    -0.26
    atum
    -0.25
    ulum
    -0.25
    èĪĴæľį
    -0.25
    æł¡
    -0.24
    åĺ§
    -0.24
    POSITIVE LOGITS
    ename
    0.25
     swell
    0.25
    걸
    0.24
     smear
    0.24
    ague
    0.24
    ç»ıåħ¸
    0.23
    Points
    0.23
    éri
    0.23
    èĢģé¾ĦåĮĸ
    0.23
    ABI
    0.23
    Activations Density 0.006%

    No Known Activations

    This feature has no known activations.