© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 44314
    Prev
    Next
    INDEX
    Explanations

    of

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on tokens carrying insults or profanities, i.e. when the text contains swear words or direct abusive language.

    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
    щини
    -0.07
    Composer
    -0.06
     заг
    -0.06
    	base
    -0.06
    πτωση
    -0.06
    	open
    -0.06
    velte
    -0.06
     şey
    -0.06
     avi
    -0.06
    ecimal
    -0.06
    POSITIVE LOGITS
    але
    0.07
    setTimeout
    0.07
     ruling
    0.07
    rias
    0.07
    ائب
    0.06
     illum
    0.06
    ido
    0.06
     предел
    0.06
    373
    0.06
    анный
    0.06
    Activations Density 0.008%

    No Known Activations