© 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. 30-TRANSCODER-HP
    7. 96432
    Prev
    Next
    INDEX
    Explanations

    of

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    è¶ĭäºİ
    -0.31
    icable
    -0.28
     oran
    -0.28
    ote
    -0.26
    æ¸ħåįķ
    -0.25
    àºĩ
    -0.25
    ç»Ĩå¿ĥ
    -0.24
     HttpRequest
    -0.24
    åºĶçĶ¨
    -0.24
    umble
    -0.23
    POSITIVE LOGITS
     biom
    0.35
     biopsy
    0.33
     biomass
    0.32
     biological
    0.31
     bios
    0.31
     bi
    0.30
     Biom
    0.29
     saliva
    0.29
     bio
    0.28
    pta
    0.28
    Activations Density 0.952%

    No Known Activations