© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 23-TRANSCODER-HP
    7. 163063
    Prev
    Next
    INDEX
    Explanations

    ach

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_23.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.23.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Seen
    -0.25
    iline
    -0.23
    olan
    -0.23
    æĪijå·²ç»ı
    -0.23
    ceeded
    -0.22
    á»ĩp
    -0.22
    entrant
    -0.22
     JNICALL
    -0.21
    throp
    -0.21
    cedures
    -0.21
    POSITIVE LOGITS
    è¡Ħ
    0.24
     |/
    0.23
     Sachs
    0.23
     pun
    0.22
    erring
    0.21
    AdapterFactory
    0.21
    exual
    0.20
     umb
    0.20
    (uuid
    0.20
    åij½
    0.20
    Activations Density 0.024%

    No Known Activations