© 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. 38839
    Prev
    Next
    INDEX
    Explanations

    a

    np_max-act · gemini-2.0-flash

    This neuron activates on the common “is a” phrasing found in company‐description boilerplate (i.e. tokens like “is” and “a” in “X is a …”).

    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
    fatal
    -0.07
     došlo
    -0.07
    adic
    -0.07
    voice
    -0.07
     Lord
    -0.06
     casos
    -0.06
    _hash
    -0.06
    Advertisements
    -0.06
    _linux
    -0.06
     plated
    -0.06
    POSITIVE LOGITS
     Activate
    0.07
     swirling
    0.07
    ेखत
    0.07
     RESOURCE
    0.06
     Along
    0.06
     Saying
    0.06
     along
    0.06
    	Node
    0.06
     roasted
    0.06
    ουμε
    0.06
    Activations Density 0.018%

    No Known Activations