© 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 EvalsExportsinterp-engineNEWAPI 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. 57517
    Prev
    Next
    INDEX
    Explanations

    it

    np_max-act · gemini-2.0-flash

    This neuron selectively activates on the consecutive tokens “it to,” as in instructional phrases like “use it to.”

    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
     con
    -0.07
    bots
    -0.07
     immature
    -0.07
     avg
    -0.07
    -Dec
    -0.07
     Stud
    -0.07
    _release
    -0.07
     LOD
    -0.07
     Davidson
    -0.07
    _female
    -0.06
    POSITIVE LOGITS
     kterém
    0.07
     použít
    0.06
     شن
    0.06
    蜜
    0.06
     them
    0.06
     ẩm
    0.06
    ست
    0.06
    ्च
    0.06
    roit
    0.06
    namespace
    0.06
    Activations Density 0.020%

    No Known Activations