© 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. 4-TRANSCODER-HP
    7. 4
    Prev
    Next
    INDEX
    Explanations

    positive descriptions

    np_max-act · gemini-2.0-flash

    words with derivational suffixes marking manner, qualities, or abstract states (e.g., adverbs/adjectival forms and nominalizations).

    oai_token-act-pair · gpt-5Triggered by @liuliujiujiu05
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_4.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.4.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    anism
    -0.25
    elig
    -0.25
    Ìģ
    -0.24
     моменÑĤ
    -0.24
    Redux
    -0.24
     tav
    -0.23
     BELOW
    -0.23
    .initializeApp
    -0.23
    fd
    -0.22
    ocs
    -0.22
    POSITIVE LOGITS
    å¸Ĥåľºä¸»ä½ĵ
    0.28
    æīĵçĿĢ
    0.26
    zar
    0.26
    @store
    0.24
    导è´Ń
    0.23
    æı£
    0.23
    åĩĨ
    0.23
    è¿Ļ家åħ¬åı¸
    0.23
    branches
    0.22
    PTS
    0.22
    Activations Density 0.021%

    No Known Activations