© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 30-TRANSCODER-HP
    7. 97053
    Prev
    Next
    INDEX
    Explanations

    de

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    untime
    -0.26
    oras
    -0.26
     sentiment
    -0.26
    ãģıãĤĮ
    -0.25
    nesty
    -0.25
    åĩĮ
    -0.25
     condom
    -0.24
    vinces
    -0.24
    nds
    -0.24
    els
    -0.24
    POSITIVE LOGITS
    å¼¹
    0.31
     hydrated
    0.30
    æĢģ
    0.27
    骨干
    0.27
    rik
    0.26
     dried
    0.26
    hydration
    0.26
    hydr
    0.26
    å¹²çĩ¥
    0.26
    Ĭ¶
    0.26
    Activations Density 0.001%

    No Known Activations