© 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
    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. 895
    Prev
    Next
    INDEX
    Explanations

    it

    np_max-act · gemini-2.0-flash

    conversational and colloquial expressions, particularly in a humorous or light-hearted context.

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

    The neuron activates on words and phrases describing hugging or affectionate physical embraces.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    detects mentions of hugging or close physical/affectionate contact between people.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    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
     industry
    -0.07
    ستان
    -0.07
     Mutex
    -0.06
    _before
    -0.06
     split
    -0.06
    ija
    -0.06
    лоб
    -0.06
     papel
    -0.06
    ró
    -0.06
     splits
    -0.06
    POSITIVE LOGITS
     procrast
    0.07
    /Game
    0.07
     spolup
    0.06
    ADMIN
    0.06
    \Session
    0.06
    Fort
    0.06
    ruary
    0.06
    _WIFI
    0.06
    Projectile
    0.06
    aterno
    0.06
    Activations Density 0.191%

    No Known Activations