© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 9326
    Prev
    Next
    INDEX
    Explanations

    sterilization

    np_max-act · gemini-2.0-flash

    This neuron detects occurrences of the root “activat/​inactivat” (e.g. activation, activated, inactivation, inactivated) and its morphological variants.

    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
    inals
    -0.07
     BST
    -0.06
     Bent
    -0.06
     Reich
    -0.06
    arel
    -0.06
    ū
    -0.06
    _FOR
    -0.06
     Steel
    -0.06
     retaliation
    -0.06
     appealed
    -0.06
    POSITIVE LOGITS
    	builder
    0.07
    .Element
    0.07
    _BOX
    0.07
    ília
    0.07
    сед
    0.06
    (Document
    0.06
     ‫
    0.06
    Added
    0.06
    ucceeded
    0.06
    setBackground
    0.06
    Activations Density 0.010%

    No Known Activations