Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 30316
    Prev
    Next
    INDEX
    Explanations

    vestib

    np_max-act · gemini-2.0-flash

    The neuron is strongly activated by terms referring to vestibular system disorders and symptoms—especially words like “vertigo,” “vestibular,” “dizziness,” “balance,” etc.

    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
     Однак
    -0.08
    اصيل
    -0.07
     Corp
    -0.06
    Indexes
    -0.06
     адже
    -0.06
     compartir
    -0.06
    持
    -0.06
    ++++++++++++++++
    -0.06
    Fax
    -0.06
     milf
    -0.06
    POSITIVE LOGITS
     vestib
    0.08
    atrib
    0.07
    Exists
    0.07
    ベル
    0.06
    ube
    0.06
     +=
    0.06
    icio
    0.06
    stuff
    0.06
     determining
    0.06
    ester
    0.06
    Activations Density 0.002%

    No Known Activations