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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 23-TRANSCODER-HP
    7. 163285
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    æľ¬è½®
    -0.29
     cigar
    -0.27
    wards
    -0.25
     comp
    -0.25
    æݼ
    -0.24
     forwards
    -0.24
    nock
    -0.24
     underestimate
    -0.24
     elevated
    -0.24
     deg
    -0.24
    POSITIVE LOGITS
     Lia
    0.29
     <![
    0.27
    abin
    0.27
     karÅŁ
    0.26
    loquent
    0.25
     \↵↵
    0.25
    Æ¡
    0.25
    ourcing
    0.25
    (Uri
    0.25
    rieved
    0.24
    Activations Density 0.016%

    No Known Activations

    This feature has no known activations.