© 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. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 23192
    Prev
    Next
    INDEX
    Explanations

    of

    np_max-act · gemini-2.0-flash

    The neuron responds to the literal token “Synthetic,” i.e. it picks out the word “Synthetic” in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    specific organic compounds used in chemical synthesis.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     tote
    -0.08
     Ownership
    -0.06
     створ
    -0.06
     thẻ
    -0.06
    phan
    -0.06
     merger
    -0.06
    abric
    -0.06
     nack
    -0.06
    .Cast
    -0.06
    mk
    -0.06
    POSITIVE LOGITS
    .responseText
    0.07
    (cs
    0.06
    (sec
    0.06
    uments
    0.06
     condo
    0.06
     linebacker
    0.06
     textStatus
    0.06
     scrapped
    0.06
    _listener
    0.06
     کاربران
    0.06
    Activations Density 0.002%

    No Known Activations