© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 12958
    Prev
    Next
    INDEX
    Explanations

    from

    np_max-act · gemini-2.0-flash

    The neuron fires on bits of text that specify actions or schedules—verbs of doing (open, resets, must be done) together with date/time indicators (October, 15, two years ago)—i.e. phrasing you’d see in instructions or event announcements.

    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
     aggravated
    -0.07
     пло
    -0.07
    (min
    -0.06
    =item
    -0.06
    菲
    -0.06
     Horn
    -0.06
    800
    -0.06
     horn
    -0.06
     đặt
    -0.06
     costly
    -0.06
    POSITIVE LOGITS
     të
    0.06
     RTWF
    0.06
     cruz
    0.06
    çuk
    0.06
     Tol
    0.06
    átel
    0.06
    щик
    0.06
    素
    0.06
     슈
    0.06
    ishly
    0.06
    Activations Density 0.342%

    No Known Activations