© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 31-TRANSCODER-HP
    7. 17782
    Prev
    Next
    INDEX
    Explanations

    action

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    åıĹ
    -0.29
    dee
    -0.28
    .Areas
    -0.27
    è¾ĵ
    -0.27
    .IntPtr
    -0.26
    rez
    -0.26
    åı¦ä¸ĢåįĬ
    -0.26
    azu
    -0.25
     è¾ĵ
    -0.25
    è°Į
    -0.25
    POSITIVE LOGITS
     actions
    0.31
    act
    0.28
    .getInteger
    0.28
    actions
    0.28
    akis
    0.27
     Actions
    0.27
    æ¾İæ¹ĥæĸ°éĹ»
    0.27
    Actions
    0.27
     action
    0.27
    .actions
    0.27
    Activations Density 0.008%

    No Known Activations