© 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. 79101
    Prev
    Next
    INDEX
    Explanations

    Invention descriptions

    np_max-act · gemini-2.0-flash

    technical descriptions of inventions or devices related to mechanical systems.

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

    This neuron specializes in spotting patent-claim language—especially the names of device parts and components (e.g. “device,” “cover,” “wall,” “connection,” “slot,” etc.) in technical/apparatus descriptions.

    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
    ++++++++
    -0.07
    542
    -0.07
     Sol
    -0.06
     six
    -0.06
    .coll
    -0.06
    SCRIPTION
    -0.06
    changed
    -0.06
    VERT
    -0.06
     N
    -0.06
    }),↵
    -0.06
    POSITIVE LOGITS
    [parent
    0.07
     بیشتری
    0.07
    WithOptions
    0.07
     Dagger
    0.06
    .surname
    0.06
    ancies
    0.06
     adulte
    0.06
     consolidate
    0.06
    よび
    0.06
     좀
    0.06
    Activations Density 0.033%

    No Known Activations