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. 130755
    Prev
    Next
    INDEX
    Explanations

    complex sentences

    np_max-act · gemini-2.0-flash

    The neuron fires on subjective evaluative words (especially sentiment-laden adjectives/adverbs expressing positive or negative judgment).

    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
    .UserId
    -0.07
     leader
    -0.07
     excursion
    -0.06
    _fig
    -0.06
     sx
    -0.06
    -area
    -0.06
     Walt
    -0.06
     goodies
    -0.06
     slightest
    -0.06
    _MI
    -0.06
    POSITIVE LOGITS
    ّت
    0.06
     tão
    0.06
     desenv
    0.06
    Scanner
    0.06
    Ë
    0.06
     equipments
    0.06
    COUNT
    0.06
     sebou
    0.06
    แพร
    0.06
     temporarily
    0.06
    Activations Density 0.155%

    No Known Activations