© 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. 95439
    Prev
    Next
    INDEX
    Explanations

    Character encoding

    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
    lim
    -0.28
     XMLHttpRequest
    -0.26
    æ§Į
    -0.25
    æ¯ĶåĪ©æŶ
    -0.24
    次
    -0.24
     interior
    -0.24
    reme
    -0.24
     coup
    -0.24
     Hammer
    -0.24
    åĽ½å®¶çº§
    -0.24
    POSITIVE LOGITS
    éģĵ
    0.72
    çļĦèĢģ
    0.44
    åºĦ
    0.41
    èĢģ
    0.41
     Äijạo
    0.38
    èİĬ
    0.36
    åĩºéģĵ
    0.35
     Dao
    0.35
     sto
    0.32
    ä½Ľ
    0.32
    Activations Density 0.883%

    No Known Activations