© 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. 31-TRANSCODER-HP
    7. 18211
    Prev
    Next
    INDEX
    Explanations

    increasing

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    éģ¥
    -0.33
     wind
    -0.31
    é£İ
    -0.30
     inert
    -0.30
    snap
    -0.29
    ::-
    -0.27
    éģ¥æݧ
    -0.27
    å¹´åĨħ
    -0.26
     gesture
    -0.26
    éļĻ
    -0.26
    POSITIVE LOGITS
     intensified
    0.32
    è¶Ĭé«ĺ
    0.30
    è¶Ĭå¤ļ
    0.30
     increasing
    0.28
     frequent
    0.28
    è¿ĶåĽŀ
    0.28
     intens
    0.28
    requent
    0.28
     increasingly
    0.28
     increased
    0.28
    Activations Density 0.936%

    No Known Activations