© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 734
    Prev
    Next
    INDEX
    Explanations

    Scientific texts

    np_max-act · gemini-2.0-flash

    The neuron activates on terminology and phrasing typical of scientific methods and experimental descriptions (e.g. “method,” “developed,” “analysis,” “equation,” “tested”).

    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
    annel
    -0.07
    imus
    -0.06
     Grimm
    -0.06
    Important
    -0.06
    ।↵
    -0.06
    lessly
    -0.06
    .↵
    -0.06
     filtro
    -0.06
    Eq
    -0.06
    itched
    -0.06
    POSITIVE LOGITS
     ipc
    0.07
     nær
    0.07
    	Session
    0.07
    $tmp
    0.07
    (expect
    0.06
    	reset
    0.06
    .fromString
    0.06
    (firstName
    0.06
    ôle
    0.06
     quiz
    0.06
    Activations Density 0.325%

    No Known Activations