© 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. 97848
    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
    MOD
    -0.26
     catching
    -0.26
    æįķæįī
    -0.24
    åĩĨåĪĻ
    -0.24
    Caught
    -0.24
    caught
    -0.24
     caught
    -0.23
     Medal
    -0.23
     Catch
    -0.23
    essler
    -0.23
    POSITIVE LOGITS
    odox
    0.31
    æŀī
    0.27
    ustry
    0.25
    èĻ±
    0.24
    æķĪ
    0.24
    urname
    0.24
     normalized
    0.24
    åĥĸ
    0.24
     nightly
    0.24
    afa
    0.24
    Activations Density 0.004%

    No Known Activations

    This feature has no known activations.