© 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. 92427
    Prev
    Next
    INDEX
    Explanations

    Professions/services

    np_max-act · gemini-2.0-flash

    The neuron activates on phrases naming specific professional roles or job functions (e.g., “sound design,” “marketing,” “play testing”).

    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
     d
    -0.07
     dow
    -0.07
    _SD
    -0.07
     Tested
    -0.06
    Convention
    -0.06
    Rank
    -0.06
     Mattis
    -0.06
    ühl
    -0.06
    _datetime
    -0.06
     обор
    -0.06
    POSITIVE LOGITS
    Trees
    0.06
    しても
    0.06
     rooft
    0.06
    _profit
    0.06
     synchronization
    0.06
     δημο
    0.06
    	admin
    0.05
     stacks
    0.05
     далеко
    0.05
     Stem
    0.05
    Activations Density 0.259%

    No Known Activations