© 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. 54500
    Prev
    Next
    INDEX
    Explanations

    Talking with someone

    np_max-act · gemini-2.0-flash

    This neuron detects phrases signalling reported speech or conversation (e.g. “I was talking,” “I said,” “I’m chatting with”).

    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
    _storage
    -0.06
    iação
    -0.06
    IN
    -0.06
    elib
    -0.06
    emphasis
    -0.06
    >()
    -0.06
    umba
    -0.06
    SetBranchAddress
    -0.06
     year
    -0.06
    山市
    -0.06
    POSITIVE LOGITS
    PRIMARY
    0.06
    conversation
    0.06
     문
    0.06
    autocomplete
    0.06
     Devon
    0.06
     Sour
    0.06
    .ok
    0.06
     восп
    0.06
     Jog
    0.06
     sou
    0.06
    Activations Density 0.033%

    No Known Activations