© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 56000
    Prev
    Next
    INDEX
    Explanations

    ian

    np_max-act · gemini-2.0-flash

    The neuron is highly activated by tokens that end with the suffix “ian,” i.e. by words like “Hamiltonian” or “Lagrangian.”

    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
    490
    -0.06
     나가
    -0.06
    [slot
    -0.06
    facet
    -0.06
     bột
    -0.06
    Scotland
    -0.06
     accountId
    -0.06
     fais
    -0.06
    culture
    -0.06
    ительное
    -0.06
    POSITIVE LOGITS
    ian
    0.11
    AN
    0.08
    IAN
    0.08
    IN
    0.08
     متن
    0.07
    -defined
    0.07
     QMainWindow
    0.07
    IPHER
    0.07
    chematic
    0.07
     controller
    0.07
    Activations Density 0.001%

    No Known Activations