© 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. 130089
    Prev
    Next
    INDEX
    Explanations

    The

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on tokens that begin new sentences or sections (i.e. the first substantive word at the start of a sentence/paragraph).

    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
     dispens
    -0.08
     Overs
    -0.07
     field
    -0.07
    Pages
    -0.07
     Toolbar
    -0.07
     recalling
    -0.06
     Town
    -0.06
     elapsed
    -0.06
    _PROPERTY
    -0.06
    header
    -0.06
    POSITIVE LOGITS
    าซ
    0.06
    _Left
    0.06
    wed
    0.06
     Andreas
    0.06
    	Function
    0.06
     onClose
    0.06
    (pass
    0.06
    ौल
    0.06
    wav
    0.06
     vyj
    0.06
    Activations Density 0.011%

    No Known Activations