© 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. 42744
    Prev
    Next
    INDEX
    Explanations

    processes, setups

    np_max-act · gemini-2.0-flash

    The neuron responds to technical nouns that name configurations, plans, or layouts (e.g. “profile,” “schedule,” “layout,” “structure,” “workflow”), marking procedural or setup descriptions.

    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
    enate
    -0.07
     IDE
    -0.07
    "})
    -0.07
    _datos
    -0.06
     类
    -0.06
     المت
    -0.06
     android
    -0.06
     فن
    -0.06
     И
    -0.06
    	params
    -0.06
    POSITIVE LOGITS
     crash
    0.06
    	inst
    0.06
     この
    0.06
    ิบ
    0.06
    oolStrip
    0.06
     propensity
    0.06
    九
    0.06
    ’est
    0.06
     Victory
    0.06
     Freem
    0.06
    Activations Density 0.055%

    No Known Activations