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. 80778
    Prev
    Next
    INDEX
    Explanations

    lung

    np_max-act · gemini-2.0-flash

    technical terms and identifiers related to programming, specifically in C++.

    oai_token-act-pair · gpt-4o-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
     Ca
    -0.07
    _rom
    -0.07
    ilt
    -0.07
     RANDOM
    -0.06
    _b
    -0.06
    ometr
    -0.06
     LAW
    -0.06
     fractured
    -0.06
     올
    -0.06
    {id
    -0.06
    POSITIVE LOGITS
    =yes
    0.07
     }
    ↵
    ↵
    ↵
    0.07
    366
    0.06
    整个
    0.06
    ใต
    0.06
     gee
    0.06
     churn
    0.06
    yt
    0.06
    velop
    0.06
    .highlight
    0.06
    Activations Density 0.625%

    No Known Activations