© 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. 96482
    Prev
    Next
    INDEX
    Explanations

    instructions, steps, procedure

    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
    omi
    -0.27
    æĿĥ
    -0.27
     suất
    -0.26
    ajas
    -0.24
     nomin
    -0.24
    ä¸ĵä¸ļçŁ¥è¯Ĩ
    -0.24
    aju
    -0.23
     passage
    -0.23
    ç°ĭ
    -0.23
    çĥ¦èºģ
    -0.23
    POSITIVE LOGITS
     instructions
    0.41
     steps
    0.35
     technique
    0.34
     procedure
    0.34
     method
    0.34
     methodology
    0.34
     algorithm
    0.34
    instructions
    0.34
    æĸ¹æ³ķ
    0.33
    æŃ¥éª¤
    0.32
    Activations Density 0.904%

    No Known Activations