© 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
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 19869
    Prev
    Next
    INDEX
    Explanations

    answers/critiques

    np_max-act · gemini-2.0-flash

    This neuron consistently activates on occurrences of the word “summary” (and closely surrounding summary‐related tokens), marking when a summary is referenced.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    factual inconsistencies in summaries compared to their source documents.

    oai_token-act-pair · gpt-4o-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
    hee
    -0.07
    lob
    -0.06
    IGN
    -0.06
    enne
    -0.06
    โปรแกรม
    -0.06
    asd
    -0.06
    рах
    -0.06
    ureen
    -0.06
    pair
    -0.06
    eneg
    -0.06
    POSITIVE LOGITS
    /comments
    0.07
    (:,:,
    0.07
    .CreateIndex
    0.07
    (sp
    0.07
     disap
    0.06
     ním
    0.06
     Lightweight
    0.06
     Adds
    0.06
     شکل
    0.06
    <?>>
    0.06
    Activations Density 0.008%

    No Known Activations