© 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. 89453
    Prev
    Next
    INDEX
    Explanations

    target datalayout specifications

    np_max-act · gemini-2.0-flash

    This neuron fires on numeric fields and colon-separated values in LLVM’s target data layout specification (e.g. the “-m:e-i64:64-f80:128…” line).

    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
     eof
    -0.07
    .edge
    -0.06
     txn
    -0.06
    InteractionEnabled
    -0.06
    endi
    -0.06
     Bulldogs
    -0.06
    ؟؟
    -0.06
    елов
    -0.05
     Uint
    -0.05
     패
    -0.05
    POSITIVE LOGITS
    asa
    0.07
    дром
    0.07
    ¨
    0.07
    SpaceItem
    0.06
    ост
    0.06
     درمان
    0.06
    vey
    0.06
    κει
    0.06
     horm
    0.06
    лением
    0.06
    Activations Density 0.004%

    No Known Activations