Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 86326
    Prev
    Next
    INDEX
    Explanations

    is

    np_max-act · gemini-2.0-flash

    This neuron activates on the word “Yes,” i.e. it detects an affirmative “Yes” response.

    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
    _artist
    -0.07
    聚
    -0.07
    /<
    -0.07
    datasets
    -0.06
     Pokémon
    -0.06
     favicon
    -0.06
    яч
    -0.06
     authService
    -0.06
    Pokemon
    -0.06
    _finish
    -0.06
    POSITIVE LOGITS
     illum
    0.07
    .gms
    0.07
     MSM
    0.06
     DROP
    0.06
     nailed
    0.06
    Ltd
    0.06
    _sur
    0.06
    是什么
    0.06
     गलत
    0.06
    иться
    0.06
    Activations Density 0.009%

    No Known Activations