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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 127169
    Prev
    Next
    INDEX
    Explanations

    question/period punctuation

    np_max-act · gemini-2.0-flash

    The neuron fires when the text is phrased as a user question—especially on question marks and question‐related words—i.e. it detects interrogative turns.

    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
    动物
    -0.07
     себя
    -0.07
     katılım
    -0.06
    	load
    -0.06
    *******↵
    -0.06
     obvykle
    -0.06
     resposta
    -0.06
     axes
    -0.06
     상세
    -0.06
    cdot
    -0.06
    POSITIVE LOGITS
     Jeans
    0.07
     Steelers
    0.07
    ifers
    0.06
     Platt
    0.06
     ges
    0.06
    OW
    0.06
     spontaneously
    0.06
     भगव
    0.06
    .getTotal
    0.06
     setTimeout
    0.06
    Activations Density 0.019%

    No Known Activations