© 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. 109459
    Prev
    Next
    INDEX
    Explanations

    matrix

    np_max-act · gemini-2.0-flash

    This neuron activates when the text mentions matrices or matrix‐style expressions (e.g. “matrix,” “begin{pmatrix},” “representation,” etc.).

    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
    Controls
    -0.07
    芙
    -0.06
     prosecuting
    -0.06
     &:
    -0.06
     afternoon
    -0.06
     tutoring
    -0.06
    Parm
    -0.06
     сопров
    -0.06
    .kernel
    -0.06
     usleep
    -0.06
    POSITIVE LOGITS
    *math
    0.06
     linestyle
    0.06
    compress
    0.06
     точ
    0.06
    ANTED
    0.06
     draw
    0.06
     ModelRenderer
    0.06
    ान
    0.06
    므로
    0.06
    .parseColor
    0.06
    Activations Density 0.008%

    No Known Activations