© 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. 95890
    Prev
    Next
    INDEX
    Explanations

    gene

    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
     untrue
    -0.29
     integrates
    -0.27
    ä¸įåħ¬å¹³
    -0.27
    æŀ¯
    -0.26
    è·Į
    -0.26
     invalidated
    -0.26
    _decay
    -0.26
    Cancel
    -0.26
    æ±½
    -0.26
    ä¹ĺ
    -0.25
    POSITIVE LOGITS
     active
    0.33
     kop
    0.29
    esium
    0.29
    åĪ¶ä½ľ
    0.29
    åĪ¶
    0.28
     Strand
    0.28
    åĪ¶éĢł
    0.28
    æĭ·
    0.28
    å¾ĪæŃ£å¸¸
    0.28
     Relation
    0.27
    Activations Density 0.888%

    No Known Activations