© 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. 98643
    Prev
    Next
    INDEX
    Explanations

    Questions and discussions

    np_max-act · gemini-2.0-flash
    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
    æĥ³äºĨæĥ³
    -0.28
    ertil
    -0.26
    Catch
    -0.25
     slashed
    -0.24
    jamin
    -0.24
     catch
    -0.24
    éģįåıĬ
    -0.24
     catching
    -0.24
    Rule
    -0.24
     Curt
    -0.24
    POSITIVE LOGITS
    çİ°åľº
    0.33
    ç´¢
    0.31
    åŃIJ
    0.27
    ç¥ŀè¯Ŀ
    0.27
    éĴ¢
    0.27
    å¹¼
    0.25
    å§ĵåIJį
    0.25
     flo
    0.25
    å¹´è½»
    0.25
    åĽ½å®¶
    0.24
    Activations Density 0.001%

    No Known Activations