© 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. 44436
    Prev
    Next
    INDEX
    Explanations

    Stopwords

    np_max-act · gemini-2.0-flash

    This neuron fires on text describing how something is done, especially the phrase “by means of” or simple uses of “by” to introduce a method or mechanism.

    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
    trap
    -0.06
     susceptible
    -0.06
    drop
    -0.06
    Already
    -0.06
     Transparency
    -0.06
    	tile
    -0.06
    113
    -0.06
    aters
    -0.06
     manuscript
    -0.06
    Link
    -0.06
    POSITIVE LOGITS
     oluyor
    0.08
    実
    0.07
    มนตร
    0.07
     portfolios
    0.06
    "${
    0.06
     sayesinde
    0.06
     soda
    0.06
    ‬↵
    0.06
     الذه
    0.06
    =\""
    0.06
    Activations Density 0.026%

    No Known Activations