© 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. 108176
    Prev
    Next
    INDEX
    Explanations

    and

    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
    imately
    -0.29
    æķĻ导
    -0.28
    çݰ代
    -0.27
    leton
    -0.26
    \:
    -0.26
    æĹĥ
    -0.26
     gamma
    -0.26
    åIJĮè¡Į
    -0.25
    plies
    -0.25
     PURE
    -0.24
    POSITIVE LOGITS
     overwritten
    0.29
    uraa
    0.27
     supers
    0.26
    æĹĹå¸ľ
    0.26
    é¦Ĵ
    0.26
     replaced
    0.26
    å±ĭ
    0.25
     ignored
    0.25
     tolerated
    0.25
    shima
    0.25
    Activations Density 0.983%

    No Known Activations