© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 56051
    Prev
    Next
    INDEX
    Explanations

    place descriptions

    np_max-act · gemini-2.0-flash

    This neuron fires on the initial capitalized tokens of an article’s subject name—i.e. the leading word(s) of a proper‐noun title or entity name.

    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
    	packet
    -0.08
    	model
    -0.08
    .bd
    -0.07
     helps
    -0.07
    _slots
    -0.07
     Neutral
    -0.07
     emits
    -0.06
    ])↵↵
    -0.06
    	types
    -0.06
     Wanna
    -0.06
    POSITIVE LOGITS
    黎
    0.06
     Кри
    0.06
    谁
    0.06
     chiều
    0.06
    _SYN
    0.06
     Zimbabwe
    0.06
     रखत
    0.06
    งต
    0.06
     різ
    0.06
     RaisePropertyChanged
    0.05
    Activations Density 0.035%

    No Known Activations