© 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. 25-TRANSCODER-HP
    7. 108571
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     bes
    -0.26
    INGTON
    -0.26
    ington
    -0.25
     mounts
    -0.25
     trainer
    -0.25
    å¹³åİŁ
    -0.24
    æĤĶ
    -0.24
    #,
    -0.24
    é«ĺæ°´å¹³
    -0.24
    Advertisements
    -0.24
    POSITIVE LOGITS
    .mit
    0.29
    лÑİ
    0.26
    åİĭåĪ¶
    0.25
    åłĩ
    0.24
     Lambda
    0.24
     dense
    0.24
    梳
    0.24
     quad
    0.24
    roc
    0.23
    ç¢İ
    0.23
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.