© 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. 108168
    Prev
    Next
    INDEX
    Explanations

    saying

    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
    为æŃ¢
    -0.29
    æ§
    -0.28
     welcome
    -0.27
    åı¦ä¸ĢåįĬ
    -0.26
     option
    -0.26
    §
    -0.24
    ivia
    -0.24
    amine
    -0.24
    uba
    -0.23
    妥
    -0.23
    POSITIVE LOGITS
    说çĿĢ
    0.50
    说å®Į
    0.45
     Saying
    0.34
    æį®æѤ
    0.32
    说åĪ°è¿ĻéĩĮ
    0.31
     saying
    0.28
    äºİæĺ¯
    0.27
    說
    0.27
    è¯Ŀ
    0.26
    ::__
    0.26
    Activations Density 0.920%

    No Known Activations