© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 73927
    Prev
    Next
    INDEX
    Explanations

    fruit

    np_max-act · gemini-2.0-flash

    This neuron activates whenever the word “fruit” (especially as a standalone title or key term) appears in the text.

    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
     terribly
    -0.08
     Gö
    -0.07
    (Log
    -0.07
    Watcher
    -0.07
     Dave
    -0.07
     McMahon
    -0.07
     Zone
    -0.07
     Nightmare
    -0.07
     Phase
    -0.07
     Sophie
    -0.07
    POSITIVE LOGITS
     fruit
    0.17
     Fruit
    0.14
     fruits
    0.10
    fruit
    0.09
    ruit
    0.09
     fruition
    0.09
    ruits
    0.08
     fruity
    0.08
     fruitful
    0.08
    trait
    0.08
    Activations Density 0.008%

    No Known Activations