© 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. 107710
    Prev
    Next
    INDEX
    Explanations

    that

    np_max-act-logits · gemini-2.0-flash
    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
    å°ĺ
    -0.32
    èĻ½è¯´
    -0.31
    å¦Ĥæŀľè¯´
    -0.31
    urator
    -0.30
    anse
    -0.28
    ueur
    -0.28
    大æĪ·
    -0.28
    quina
    -0.27
    eur
    -0.25
    ince
    -0.25
    POSITIVE LOGITS
    ä»»ä½ķ
    0.43
    ä»»æĦı
    0.41
    æľīä»»ä½ķ
    0.41
     any
    0.37
    ä»»ä½ķä¸Ģ个
    0.36
     anybody
    0.35
    Any
    0.34
     ANY
    0.34
    any
    0.34
     anyone
    0.33
    Activations Density 1.059%

    No Known Activations