© 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. 96091
    Prev
    Next
    INDEX
    Explanations

    prediction

    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
    ices
    -0.27
    éŃĶ
    -0.27
    è¡¥
    -0.26
    èĹ»
    -0.26
    Ob
    -0.26
     front
    -0.26
    á»ĥn
    -0.26
    .obs
    -0.26
    æµģéĩı
    -0.26
    å²Ĺ
    -0.25
    POSITIVE LOGITS
    æľĪä¸Ń
    0.28
    人éĹ´
    0.27
    atto
    0.27
    igung
    0.26
     asserts
    0.26
    tÃŃ
    0.25
    éĩİçĶŁ
    0.25
    ERE
    0.24
    åľ¨è¿ĩ渡
    0.24
    deen
    0.24
    Activations Density 0.001%

    No Known Activations