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

    Neuronpedia

    Jacobian LensNEW
    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. 56506
    Prev
    Next
    INDEX
    Explanations

    debate

    np_max-act · gemini-2.0-flash

    discussions of political debates and their implications on candidates' performances.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron detects mentions of “debate,” strongly activating on the word “debate” (and closely related debate‐context tokens).

    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
     눈
    -0.06
     Посилання
    -0.06
    _mx
    -0.06
    を見
    -0.06
     throat
    -0.06
    _call
    -0.06
    	mp
    -0.06
     кла
    -0.06
     صف
    -0.06
     Shipping
    -0.06
    POSITIVE LOGITS
     OpCode
    0.07
     rhetoric
    0.07
    @ResponseBody
    0.07
     đèn
    0.06
    (withId
    0.06
    antasy
    0.06
     своим
    0.06
    forEach
    0.06
     obvyk
    0.06
     первой
    0.06
    Activations Density 0.005%

    No Known Activations