© 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. 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. 58597
    Prev
    Next
    INDEX
    Explanations

    this

    np_max-act · gemini-2.0-flash

    The neuron activates on the “assistant” speaker tags in the conversation headers, identifying when the assistant is speaking.

    oai_token-act-pair · o4-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
     Glover
    -0.07
    ComboBox
    -0.07
     shopping
    -0.06
    ennial
    -0.06
    412
    -0.06
    576
    -0.06
     shade
    -0.06
     stale
    -0.06
    Poll
    -0.06
    済
    -0.06
    POSITIVE LOGITS
     hüküm
    0.07
    ัพ
    0.06
     iVar
    0.06
     다음
    0.06
    (policy
    0.06
    	names
    0.06
    (月
    0.06
     Bau
    0.06
    liğ
    0.06
    subclass
    0.06
    Activations Density 0.047%

    No Known Activations