© 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. 9500
    Prev
    Next
    INDEX
    Explanations

    Character descriptions/creation

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of the word “character” (or its French equivalent “personnage”) in prompts or discussion.

    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
     بت
    -0.06
    但
    -0.06
    _ON
    -0.06
     warranty
    -0.06
    ｜
    -0.06
    mediately
    -0.06
    burgh
    -0.06
     paranormal
    -0.05
     trolls
    -0.05
    κα
    -0.05
    POSITIVE LOGITS
     interface
    0.07
     useNewUrlParser
    0.07
    kses
    0.07
     yes
    0.07
    変更
    0.06
     ArrayAdapter
    0.06
     seriously
    0.06
    _Bool
    0.06
    InitStruct
    0.06
     Sol
    0.06
    Activations Density 0.030%

    No Known Activations