© 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. 126717
    Prev
    Next
    INDEX
    Explanations

    ,

    np_max-act · gemini-2.0-flash

    The neuron responds to structural or formatting markers—punctuation (colons, commas, quotes, parentheses), numeric tokens (years, numbers), and similar layout cues.

    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
    -db
    -0.06
    Down
    -0.06
     Nam
    -0.06
    Containers
    -0.06
    Mic
    -0.06
    Instr
    -0.06
     looming
    -0.06
     Antique
    -0.06
     asshole
    -0.06
     Meh
    -0.06
    POSITIVE LOGITS
     verdade
    0.07
    ность
    0.07
     frac
    0.06
     peu
    0.06
    ็บ
    0.06
    عة
    0.06
    	Connection
    0.06
    quad
    0.06
    ัตร
    0.06
    ۱۹۷
    0.06
    Activations Density 0.184%

    No Known Activations