© 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 (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 89838
    Prev
    Next
    INDEX
    Explanations

    Words ending in "us"

    np_max-act · gemini-2.0-flash

    The neuron activates on proper names (especially classical or mythological place/person names) ending in the “-ius” (or similar Latin/Greek) suffix.

    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
     e
    -0.07
    ’te
    -0.07
    'e
    -0.07
    Free
    -0.07
     DE
    -0.07
     en
    -0.07
    'E
    -0.07
     JE
    -0.07
    jící
    -0.06
     Free
    -0.06
    POSITIVE LOGITS
    us
    0.16
    ium
    0.13
    UM
    0.13
    um
    0.12
    US
    0.12
    ius
    0.11
    inus
    0.11
    enus
    0.11
    anium
    0.11
    acus
    0.11
    Activations Density 0.066%

    No Known Activations