Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 86144
    Prev
    Next
    INDEX
    Explanations

    exceptional

    np_max-act · gemini-2.0-flash

    The neuron activates specifically on the adjective “exceptional,” flagging instances of that word in the text.

    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
    .remove
    -0.06
    _REFRESH
    -0.06
    thumb
    -0.06
    ='\
    -0.06
     ante
    -0.06
    		↵		↵		↵
    -0.06
    	boolean
    -0.06
    _BUSY
    -0.06
    	UP
    -0.06
    -coded
    -0.06
    POSITIVE LOGITS
     exceptional
    0.12
     exceptionally
    0.08
     Occasionally
    0.07
    ssc
    0.07
     supernatural
    0.07
    جی
    0.07
     شه
    0.07
     unique
    0.07
    秀
    0.07
    edin
    0.07
    Activations Density 0.008%

    No Known Activations