Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 43590
    Prev
    Next
    INDEX
    Explanations

    numbers

    np_max-act · gemini-2.0-flash

    patient identifiers and medical report details.

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

    The neuron is highly sensitive to numeric identifiers (especially accession‐style numbers) in the document.

    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
    OutputStream
    -0.07
    InputStream
    -0.07
    gba
    -0.07
    -responsive
    -0.06
    -type
    -0.06
     ولد
    -0.06
     Ка
    -0.06
    _rg
    -0.06
     огром
    -0.06
    씨
    -0.06
    POSITIVE LOGITS
     LATIN
    0.07
     Testing
    0.07
                                                                                               
    0.06
     طول
    0.06
    .Enable
    0.06
     verificar
    0.06
     humorous
    0.06
     трь
    0.06
    agger
    0.06
     shaved
    0.06
    Activations Density 0.002%

    No Known Activations