© 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. 57057
    Prev
    Next
    INDEX
    Explanations

    irre

    np_max-act · gemini-2.0-flash

    references to legal terms related to divorce and family law.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron detects the “irre–” prefix in words (e.g. irreversible, irreparable) indicating irreversibility or irreparable damage.

    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
     Lift
    -0.06
     condem
    -0.06
     fed
    -0.06
     jab
    -0.06
    	mat
    -0.06
    student
    -0.06
    谈
    -0.06
     جون
    -0.06
    Bed
    -0.06
    ento
    -0.06
    POSITIVE LOGITS
     irre
    0.11
     Ir
    0.10
     Irr
    0.08
     ir
    0.07
     unusually
    0.07
    ayıp
    0.07
    _IR
    0.07
    Ir
    0.07
     irrig
    0.07
     FIRE
    0.07
    Activations Density 0.006%

    No Known Activations