© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 80309
    Prev
    Next
    INDEX
    Explanations

    comma

    np_max-act · gemini-2.0-flash

    detailed descriptors for artistic or imaginative imagery.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    Neuron 1: It fires on present‐participle verbs (words ending in “-ing”). Neuron 2: It detects medical or bodily conditions (e.g. “arrest,” “vertigo”). Neuron 3: It responds to expressions of unity or togetherness. Neuron 4: It picks out high-end image‐style and quality descriptors (photorealistic, lighting, resolution, 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
    ανδ
    -0.07
     onslaught
    -0.06
     knockout
    -0.06
    gest
    -0.06
    Double
    -0.06
     Buff
    -0.06
    ropped
    -0.06
     Dip
    -0.06
    	Set
    -0.06
     Call
    -0.06
    POSITIVE LOGITS
    _userdata
    0.07
     getY
    0.07
     Pods
    0.07
    >tagger
    0.07
     onay
    0.07
     боль
    0.06
     首页第
    0.06
    .Reporting
    0.06
    ?('
    0.06
     삶
    0.06
    Activations Density 0.020%

    No Known Activations