© 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. 108726
    Prev
    Next
    INDEX
    Explanations

    organizations and their actions

    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
    解çŃĶ
    -0.29
    å®¡æł¸
    -0.29
    å®¡æŁ¥
    -0.27
    ä¼ļ被
    -0.27
    被
    -0.27
    abble
    -0.26
    Checking
    -0.26
    好çļĦ
    -0.26
    æ¶īåıĬåĪ°
    -0.25
    iw
    -0.25
    POSITIVE LOGITS
     nær
    0.27
     therefore
    0.27
    InSection
    0.27
    _kv
    0.27
    orian
    0.26
    .setItem
    0.26
     decided
    0.26
     bracket
    0.25
    ataka
    0.25
    ä¹ĭéĹ´
    0.25
    Activations Density 0.896%

    No Known Activations