© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 7-RESID-POST-AA
    11. 63402
    Prev
    Next
    INDEX
    Explanations

    ky

    np_max-act · gemini-2.0-flash

    named entities, especially capitalized multi‑token proper names of people, places, institutions, and companies (often including titles or identifiers).

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_7/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.7.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Fro
    -0.07
    orical
    -0.07
     Á
    -0.07
     aph
    -0.06
    ::::
    -0.06
    _expected
    -0.06
     cherche
    -0.06
     DET
    -0.06
     hrom
    -0.06
     قادر
    -0.06
    POSITIVE LOGITS
    .act
    0.08
    ...↵↵↵↵
    0.07
     modify
    0.07
    820
    0.06
     sdl
    0.06
    ُه
    0.06
     Techniques
    0.06
    ิษ
    0.06
     Phó
    0.06
     initiating
    0.06
    Activations Density 0.739%

    No Known Activations