© 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. 25-TRANSCODER-HP
    7. 107624
    Prev
    Next
    INDEX
    Explanations

    in/of

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     signatures
    -0.26
    WORDS
    -0.26
    çļĦ表æĥħ
    -0.25
    æĶ¶è´¹
    -0.25
    好å¥ĩå¿ĥ
    -0.24
    å®ŀè·µä¸Ń
    -0.24
     royalties
    -0.24
    æł¸ç®Ĺ
    -0.24
    adients
    -0.24
    urile
    -0.24
    POSITIVE LOGITS
    rous
    0.26
    Mui
    0.26
    ìĨĮ
    0.25
    磼
    0.24
    WithTag
    0.24
    Filtered
    0.24
     wholesome
    0.24
     stripped
    0.24
    dap
    0.23
     bonded
    0.23
    Activations Density 0.932%

    No Known Activations