© 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 EvalsExportsinterp-engineNEWAPI 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. 46667
    Prev
    Next
    INDEX
    Explanations

    ey

    np_max-act · gemini-2.0-flash

    specific descriptive adjectives for texture or consistency.

    oai_token-act-pair · gpt-4oTriggered by @tcai

    The neuron strongly activates on short descriptive multi‐word noun phrases or compounds—especially adjective–noun or noun–with–noun constructions (e.g. “gooey thick glob,” “migraines with auras,” “pizza parlour crunch,” “pink things”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    words and phrases related to beauty and personal care products.

    oai_token-act-pair · gpt-4o-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
     곧
    -0.07
     printk
    -0.06
     datings
    -0.06
    驶
    -0.06
     مط
    -0.06
    _stderr
    -0.06
     할인
    -0.06
    厅
    -0.06
     khuôn
    -0.06
     nhất
    -0.06
    POSITIVE LOGITS
     underwent
    0.07
    	void
    0.07
    Found
    0.07
    най
    0.06
     Percentage
    0.06
     MOV
    0.06
    Par
    0.06
    .permissions
    0.06
     argues
    0.06
     range
    0.06
    Activations Density 0.044%

    No Known Activations