© 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. 126931
    Prev
    Next
    INDEX
    Explanations

    Analysis/Observation

    np_max-act · gemini-2.0-flash

    This neuron fires on analytical framing words in hypothetical or conditional expressions (e.g. “we look,” “one takes,” “we consider”) that introduce examples or evidence.

    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
     твор
    -0.08
     yaşında
    -0.07
    866
    -0.06
    ским
    -0.06
     Dick
    -0.06
    מ
    -0.06
    าคา
    -0.06
    kova
    -0.06
    rette
    -0.06
     smart
    -0.06
    POSITIVE LOGITS
     regarded
    0.07
    (se
    0.06
     doğrult
    0.06
    /connect
    0.06
    	tr
    0.06
    ="-
    0.06
    	image
    0.06
    ­i
    0.06
    illustr
    0.06
    >v
    0.06
    Activations Density 0.032%

    No Known Activations