© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 128313
    Prev
    Next
    INDEX
    Explanations

    improvements

    np_max-act · gemini-2.0-flash

    This neuron detects purposive or goal-oriented language—especially verbs and phrases (like “to,” “enhance,” “reduce,” “improve,” “avoid,” etc.) that introduce stated objectives or improvements in a technical context.

    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
    ura
    -0.07
    uyu
    -0.06
    Hell
    -0.06
    ELL
    -0.06
    .layouts
    -0.06
    boarding
    -0.06
    Harry
    -0.06
     основном
    -0.06
     yourselves
    -0.06
    において
    -0.06
    POSITIVE LOGITS
    quiv
    0.07
    toContain
    0.06
    áj
    0.06
     persever
    0.06
     coupe
    0.06
     bols
    0.06
    bris
    0.06
     đế
    0.06
    osci
    0.06
    Reason
    0.06
    Activations Density 0.054%

    No Known Activations