© 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. 914
    Prev
    Next
    INDEX
    Explanations

    numbers and identifiers

    np_max-act · gemini-2.0-flash

    This neuron responds to date- and time-related header fields (e.g. “Date,” “Last-Modified,” “If-Modified-Since”) in HTTP metadata.

    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
     Ка
    -0.07
     διά
    -0.07
    (inputs
    -0.07
    xd
    -0.06
     anticipated
    -0.06
     Charg
    -0.06
     Republic
    -0.06
     attorneys
    -0.06
     interfering
    -0.06
    ازی
    -0.06
    POSITIVE LOGITS
     genuine
    0.07
    .showToast
    0.06
    GestureRecognizer
    0.06
     Tac
    0.06
     reluct
    0.06
    ského
    0.06
     RTAL
    0.06
    -loving
    0.06
     becer
    0.06
     british
    0.06
    Activations Density 1.346%

    No Known Activations