© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 23-TRANSCODER-HP
    7. 162965
    Prev
    Next
    INDEX
    Explanations

    adverbs/conjunctions

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_23.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.23.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     accurate
    -0.33
     much
    -0.31
     pronounced
    -0.31
    doi
    -0.30
    é«ĺçļĦ
    -0.30
     diverse
    -0.28
    much
    -0.28
    å¤ļåľ°
    -0.28
    大çļĦ
    -0.28
    æ¿ĢçĥĪçļĦ
    -0.28
    POSITIVE LOGITS
     very
    0.33
    æ¯Ķè¾ĥ
    0.29
    å¾Ī
    0.29
    éĿŀ常ãģ«
    0.26
    éĿŀ常
    0.26
    æĪijå¾Ī
    0.26
    å¾Īæ·±
    0.26
    izia
    0.26
     ganz
    0.26
    steel
    0.26
    Activations Density 0.067%

    No Known Activations