© 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
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 74698
    Prev
    Next
    INDEX
    Explanations

    H89 (protein kinase inhibitor)

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of small‐molecule inhibitor names (e.g. kinase inhibitors like H-89, KT5720, SB203580, U0126, IBMX, calphostin C, 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
    589
    -0.07
    ongoose
    -0.07
     wherever
    -0.06
    asu
    -0.06
    Associated
    -0.06
     ironically
    -0.06
    ریک
    -0.06
    _controller
    -0.06
     Armen
    -0.06
     Buk
    -0.06
    POSITIVE LOGITS
    .linear
    0.07
    ThreadId
    0.07
     наблюд
    0.07
     nového
    0.06
    .Inv
    0.06
     gerçekten
    0.06
     vzdál
    0.06
    Про
    0.06
    érience
    0.06
    Such
    0.06
    Activations Density 0.008%

    No Known Activations