Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 1000
    Prev
    Next
    INDEX
    Explanations

    term

    np_max-act · gemini-2.0-flash

    The neuron activates on mentions of preterm birth or prematurity.

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

    identifying and highlighting risks, complications, and outcomes associated with preterm birth.

    oai_token-act-pair · gpt-5-nanoTriggered by @tcai
    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
     equitable
    -0.07
     Cleans
    -0.06
    _ud
    -0.06
     onions
    -0.06
    kö
    -0.06
     grazing
    -0.06
     shipment
    -0.06
    anske
    -0.06
    (dic
    -0.06
     techniques
    -0.06
    POSITIVE LOGITS
    _ALARM
    0.07
    .prompt
    0.07
     turns
    0.07
    _RESOLUTION
    0.07
     specializing
    0.07
    tuğ
    0.06
     outro
    0.06
     Đào
    0.06
    フト
    0.06
    PTS
    0.06
    Activations Density 0.002%

    No Known Activations