© 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. 68345
    Prev
    Next
    INDEX
    Explanations

    uppercase letters

    np_max-act · gemini-2.0-flash

    The neuron fires on contiguous character runs in descending (reverse) order, e.g. “zyx…”, “UTSRQP…”, “987654321,” and similar backwards letter or digit sequences.

    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
     checkBox
    -0.07
     Hydraulic
    -0.07
    Offer
    -0.07
     Acts
    -0.07
    	push
    -0.07
     wildcard
    -0.06
    EMS
    -0.06
     бок
    -0.06
    AllowAnonymous
    -0.06
    ัวเอง
    -0.06
    POSITIVE LOGITS
     undercover
    0.06
    [C
    0.06
    др
    0.06
     fundraiser
    0.06
    .constraints
    0.06
    Fraction
    0.06
     초기
    0.06
     الخط
    0.06
    向
    0.05
    (sprintf
    0.05
    Activations Density 0.007%

    No Known Activations