© 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. 41348
    Prev
    Next
    INDEX
    Explanations

    possessive pronouns

    np_max-act · gemini-2.0-flash

    This neuron never activates on ordinary words or punctuation—essentially it’s a “dead” neuron that only (theoretically) responds to special control or protocol tokens, not to normal document content.

    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
    Start
    -0.07
    006
    -0.07
     Glock
    -0.06
     rede
    -0.06
    orch
    -0.06
     Борис
    -0.06
     Ars
    -0.06
     Cumberland
    -0.06
    IOC
    -0.06
    祝
    -0.06
    POSITIVE LOGITS
    mayan
    0.07
     espaço
    0.06
    JI
    0.06
    別
    0.06
     millions
    0.06
    DIM
    0.06
     другими
    0.06
     ड
    0.06
    金属
    0.06
     Danielle
    0.06
    Activations Density 0.025%

    No Known Activations