© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 15-RESID-POST-AA
    11. 38011
    Prev
    Next
    INDEX
    Explanations

    punctuation and common words

    np_max-act · gemini-2.0-flash

    Tokens that occur at the start of a new paragraph or document segment (i.e., paragraph-initial tokens).

    oai_token-act-pair · gpt-5-mini
    Triggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_15/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     Vocabulary
    -0.06
     mushroom
    -0.06
    {}↵↵
    -0.06
    	bytes
    -0.06
     utter
    -0.06
    	trigger
    -0.06
     metros
    -0.05
    　
    -0.05
     ****************************************
    -0.05
    .coeff
    -0.05
    POSITIVE LOGITS
    сі
    0.07
    ाक
    0.07
    Manage
    0.07
    =<
    0.06
     mnohem
    0.06
    ов
    0.06
    ildiği
    0.06
     ét
    0.06
    른
    0.06
    resources
    0.06
    Activations Density 0.000%

    No Known Activations