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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 89845
    Prev
    Next
    INDEX
    Explanations

    

    np_max-act · gemini-2.0-flash

    The neuron detects the presence of a non-printing byte-order-mark (BOM) character (U+FEFF) at the start of a text.

    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
    ign
    -0.08
    IGN
    -0.08
     Мин
    -0.07
    tol
    -0.07
    lararası
    -0.06
     бол
    -0.06
    ποί
    -0.06
    dden
    -0.06
    jur
    -0.06
    :|
    -0.06
    POSITIVE LOGITS
     strangely
    0.07
    ='$
    0.06
    letcher
    0.06
     characteristics
    0.06
     this
    0.06
    (withIdentifier
    0.06
     Plaintiff
    0.06
     texture
    0.06
     accessing
    0.06
     sạch
    0.06
    Activations Density 0.000%

    No Known Activations