© 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. 25736
    Prev
    Next
    INDEX
    Explanations

    surge

    np_max-act · gemini-2.0-flash

    surge-related terms and concepts in medical contexts.

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

    The neuron activates whenever it sees the word (or subword) “surge,” regardless of context.

    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
    ätt
    -0.07
    犯
    -0.06
     Coat
    -0.06
    _PROVID
    -0.06
    leans
    -0.06
     formato
    -0.06
    iving
    -0.06
     labels
    -0.06
    ctor
    -0.06
    REMOVE
    -0.06
    POSITIVE LOGITS
     surge
    0.11
     resurgence
    0.08
     Suppose
    0.07
     Pulse
    0.07
    urge
    0.07
     Surge
    0.07
    eníze
    0.07
     imb
    0.07
     Česko
    0.07
    /source
    0.07
    Activations Density 0.003%

    No Known Activations