© 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. 108983
    Prev
    Next
    INDEX
    Explanations

    code snippets

    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
     worry
    -0.48
     focus
    -0.43
     rear
    -0.41
    -focus
    -0.41
     worries
    -0.40
    æĹłå¿§
    -0.39
    å®īå¿ĥ
    -0.37
    Focus
    -0.37
     concern
    -0.36
     peace
    -0.36
    POSITIVE LOGITS
    icits
    0.34
    kest
    0.30
    imir
    0.28
    oking
    0.28
    acity
    0.27
    меÑĤ
    0.26
    ictim
    0.25
     mocked
    0.24
    综åIJĪåĪ©çĶ¨
    0.24
    é¦Ĺ
    0.24
    Activations Density 0.973%

    No Known Activations