© 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. 130091
    Prev
    Next
    INDEX
    Explanations

    Questions

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric tokens—especially standalone numbers or decimal values (e.g. answer‐choice scores, counts, or question/option labels).

    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
     darker
    -0.07
    .root
    -0.07
    .splice
    -0.07
     Gdk
    -0.07
     plá
    -0.07
    ibile
    -0.07
     prove
    -0.07
    izzlies
    -0.07
    .add
    -0.07
     elves
    -0.06
    POSITIVE LOGITS
     Symptoms
    0.07
     �
    0.06
    87
    0.06
     Crushers
    0.06
    terms
    0.05
     incontro
    0.05
    Tim
    0.05
    Hibernate
    0.05
    ึ่
    0.05
    -secret
    0.05
    Activations Density 0.005%

    No Known Activations