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. 85932
    Prev
    Next
    INDEX
    Explanations

    prediction accuracy

    np_max-act · gemini-2.0-flash

    The neuron flags words and phrases that discuss the accuracy, reliability, or performance of predictions or forecasts.

    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
     promotion
    -0.07
    養
    -0.06
     ```↵
    -0.06
    .padding
    -0.06
     Howe
    -0.06
     한번
    -0.06
     forwards
    -0.06
    -max
    -0.06
    Freq
    -0.06
     overlaps
    -0.06
    POSITIVE LOGITS
    ดาห
    0.07
    _framework
    0.07
    -deals
    0.06
     $_[
    0.06
    _APP
    0.06
    _logs
    0.06
     au
    0.06
    _relation
    0.06
    _hits
    0.06
     Pussy
    0.06
    Activations Density 0.022%

    No Known Activations