© 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. 38812
    Prev
    Next
    INDEX
    Explanations

    not

    np_max-act · gemini-2.0-flash

    This neuron detects negation cues—words like “not,” “no,” “didn’t,” and related negative expressions.

    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
     counterfeit
    -0.07
     सन
    -0.07
     ภาษ
    -0.06
    imbabwe
    -0.06
    adle
    -0.06
     Kurd
    -0.06
    .area
    -0.06
     trays
    -0.06
    Ye
    -0.06
    .Mask
    -0.06
    POSITIVE LOGITS
    -react
    0.07
    743
    0.07
     inter
    0.07
    cess
    0.06
     non
    0.06
     avoiding
    0.06
     Complaint
    0.06
    ,'']]],↵
    0.06
    vf
    0.06
     abs
    0.06
    Activations Density 0.060%

    No Known Activations