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

    Neuronpedia

    Jacobian LensNEW
    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. 57303
    Prev
    Next
    INDEX
    Explanations

    fight scenes

    np_max-act · gemini-2.0-flash

    This neuron detects special structural tokens marking document boundaries or metadata (e.g. start/end headers, text‐ID tags).

    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
    led
    -0.07
     zahl
    -0.07
    bubble
    -0.06
    ined
    -0.06
     pandas
    -0.06
    Targets
    -0.06
    append
    -0.06
    ále
    -0.06
    ثمان
    -0.06
     있었
    -0.06
    POSITIVE LOGITS
     dll
    0.07
    <dd
    0.06
     جنسی
    0.06
    震
    0.06
    ，可以
    0.06
     पस
    0.06
    著
    0.06
    ταση
    0.06
     Comment
    0.06
     Business
    0.06
    Activations Density 0.034%

    No Known Activations