© 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 (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 90675
    Prev
    Next
    INDEX
    Explanations

    pathogens

    np_max-act · gemini-2.0-flash

    The neuron is selectively activating on terms related to pathogens and immune‐defense contexts (e.g., “pathogens,” “infectious,” “against,” “defending”).

    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
     فه
    -0.06
    iances
    -0.06
     сам
    -0.06
     запис
    -0.06
    不安
    -0.06
     pcs
    -0.06
     tourist
    -0.06
     lying
    -0.06
    _lost
    -0.06
     اسپ
    -0.06
    POSITIVE LOGITS
     authDomain
    0.08
     Monsters
    0.07
    _,↵
    0.07
    chin
    0.06
     IllegalStateException
    0.06
     carbohydrate
    0.06
    _SCL
    0.06
    Danger
    0.06
    START
    0.06
    unct
    0.06
    Activations Density 0.015%

    No Known Activations