© 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. 44527
    Prev
    Next
    INDEX
    Explanations

    Impact, quality, efficiency

    np_max-act · gemini-2.0-flash

    The neuron fires on abstract, multi‐syllabic nouns or adjectives—especially technical terms denoting qualities, functions, or effects (e.g. convenience, efficiency, seriousness, technology, position).

    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
     Emperor
    -0.07
     bucket
    -0.07
    どこ
    -0.06
    окрема
    -0.06
    ，是
    -0.06
    (port
    -0.06
    	sort
    -0.06
    톡
    -0.06
    ,"
    -0.06
     phương
    -0.06
    POSITIVE LOGITS
    ystals
    0.07
     CGRect
    0.06
     alloy
    0.06
     asm
    0.06
    relation
    0.06
    oldt
    0.06
    -written
    0.06
     Nowadays
    0.06
    digital
    0.06
     segreg
    0.05
    Activations Density 0.108%

    No Known Activations