© 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 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. 56855
    Prev
    Next
    INDEX
    Explanations

    have

    np_max-act · gemini-2.0-flash

    This neuron responds to the verb “have” when it’s used to express the user’s intent (e.g. “trying to have…,” “want to have…”).

    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
    tick
    -0.08
     points
    -0.07
     mirrors
    -0.07
     lines
    -0.07
     rotor
    -0.07
    ripp
    -0.06
     TEST
    -0.06
    IM
    -0.06
     moons
    -0.06
     km
    -0.06
    POSITIVE LOGITS
     Have
    0.10
     having
    0.10
     Having
    0.09
    having
    0.08
    HAVE
    0.08
    Having
    0.08
    Have
    0.08
     ハ
    0.07
    ToAdd
    0.07
     Danh
    0.07
    Activations Density 0.052%

    No Known Activations