© 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. 30-TRANSCODER-HP
    7. 95443
    Prev
    Next
    INDEX
    Explanations

    emotional state

    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
    Not in Any Lists

    No Comments

    Negative Logits
    icl
    -0.28
    iciel
    -0.26
    enas
    -0.26
    riel
    -0.26
    ITU
    -0.25
    好ä¸į好
    -0.25
    usal
    -0.24
    ä¼ļéķ¿
    -0.24
    空æ°ĶåĩĢåĮĸ
    -0.23
    ordial
    -0.23
    POSITIVE LOGITS
     that
    0.49
     at
    0.44
    that
    0.41
    seeing
    0.40
     seeing
    0.40
     learn
    0.40
    è§ģåĪ°
    0.39
     hearing
    0.37
    _learn
    0.37
    çľĭåĪ°
    0.36
    Activations Density 0.031%

    No Known Activations