© 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. 58839
    Prev
    Next
    INDEX
    Explanations

    narratives

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on the word “to” when it’s introducing an infinitive (e.g. “to find,” “to wonder”), i.e. marking “to + verb” constructions.

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

    conversations or dialogues between characters.

    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
    Contains
    -0.07
    Tour
    -0.07
     Class
    -0.06
     eSports
    -0.06
    Na
    -0.06
     baskets
    -0.06
    Layout
    -0.06
    461
    -0.06
     sponsorship
    -0.06
    Ma
    -0.06
    POSITIVE LOGITS
     bağlı
    0.07
    (protocol
    0.07
    isicing
    0.07
     calming
    0.07
    enerative
    0.06
    -factor
    0.06
     '/');↵
    0.06
     rational
    0.06
    odes
    0.06
    /firebase
    0.06
    Activations Density 0.010%

    No Known Activations