© 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. 124978
    Prev
    Next
    INDEX
    Explanations

    red

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of an elderly woman (e.g., “old” + “lady”).

    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
    _Header
    -0.08
     Commissioners
    -0.07
    획
    -0.07
    rogen
    -0.07
     бук
    -0.07
     reasoned
    -0.07
     erreur
    -0.06
    ilinear
    -0.06
    	Copyright
    -0.06
    Perl
    -0.06
    POSITIVE LOGITS
    SDK
    0.06
     selectively
    0.06
    _VOL
    0.06
     Converted
    0.06
    Posts
    0.06
    ْد
    0.06
     cumpl
    0.06
     Publish
    0.06
    ía
    0.05
    μεν
    0.05
    Activations Density 0.049%

    No Known Activations