© 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. 11-RESID-POST-AA
    11. 58695
    Prev
    Next
    INDEX
    Explanations

    Internet discussion snippets

    np_max-act · gemini-2.0-flash

    discussion and suggestions related to protective measures for outdoor items, particularly boats.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron selectively responds to tokens containing the common English/chemical suffixes “-er” (e.g. eraser, Don’t, wouldn’t) and “-one” (e.g. acetone).

    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
    HOST
    -0.06
     Mini
    -0.06
     Hack
    -0.06
     Thus
    -0.06
    unci
    -0.06
    DAY
    -0.06
    .fast
    -0.06
    anceled
    -0.06
     двух
    -0.06
    ήμερα
    -0.06
    POSITIVE LOGITS
     �
    0.07
    isContained
    0.07
     अप
    0.06
    ikip
    0.06
     самов
    0.06
     accompanying
    0.06
    Ltd
    0.06
    (any
    0.06
     التع
    0.06
    단체
    0.06
    Activations Density 0.115%

    No Known Activations