© 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. 107625
    Prev
    Next
    INDEX
    Explanations

    ending or final action

    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
    ä¿Ŀ
    -0.31
    áºŃt
    -0.28
    è¹Ń
    -0.28
    azed
    -0.28
    warz
    -0.26
    ãģ¾
    -0.26
     abide
    -0.25
    æľŁåĨħ
    -0.25
    variation
    -0.25
    eros
    -0.24
    POSITIVE LOGITS
    èĩ´åij½
    0.43
     final
    0.37
    (final
    0.36
     finishing
    0.36
     clin
    0.36
    æľĢç»Ī
    0.35
    -final
    0.35
     straw
    0.35
     finally
    0.33
    æľĢçµĤ
    0.33
    Activations Density 1.065%

    No Known Activations