© 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. 96731
    Prev
    Next
    INDEX
    Explanations

    consistency

    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
    æĺ¶
    -0.26
    uggle
    -0.24
     toggle
    -0.24
    æĸĿ
    -0.24
    айн
    -0.24
    侮辱
    -0.23
    iÄħ
    -0.23
    ius
    -0.23
    åŃĢ
    -0.23
    æĬķ
    -0.23
    POSITIVE LOGITS
     consistent
    2.02
    consistent
    1.67
     consistency
    1.59
    ä¸Ģèĩ´
    1.58
     Cons
    1.43
     inconsistent
    1.41
    ä¸Ģèĩ´æĢ§
    1.37
    Cons
    1.21
     inconsistency
    1.15
    -cons
    1.14
    Activations Density 1.182%

    No Known Activations