© 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. 89395
    Prev
    Next
    INDEX
    Explanations

    personal news

    np_max-act · gemini-2.0-flash

    This neuron fires on sentence- or section-initial transition words or phrases (e.g. “In,” “On,” “But,” “Also,” “I’ve”) that signal a new thought or topic shift.

    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
    кол
    -0.06
    -strong
    -0.06
     (>
    -0.06
    -training
    -0.06
    -th
    -0.06
    ',{
    -0.06
    vic
    -0.06
    стро
    -0.06
    /fonts
    -0.06
    elan
    -0.06
    POSITIVE LOGITS
    .GetAxis
    0.07
     OSError
    0.06
     QE
    0.06
     nya
    0.06
     giver
    0.06
    HeaderInSection
    0.06
     clase
    0.06
     clientId
    0.06
     مواطنة
    0.06
    meet
    0.06
    Activations Density 0.028%

    No Known Activations