© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 90828
    Prev
    Next
    INDEX
    Explanations

    Cant

    np_max-act · gemini-2.0-flash

    The neuron spikes on substrings of material or chemical names—most notably the “poly” and “ester” fragments of “polyester.”

    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
    7
    -0.08
    000
    -0.08
    صد
    -0.07
    ९
    -0.07
    6
    -0.07
    ४
    -0.07
    8
    -0.07
    ۸
    -0.07
    /y
    -0.07
    -analytics
    -0.07
    POSITIVE LOGITS
     provoke
    0.07
     Claus
    0.06
     gracious
    0.06
     popup
    0.06
     coach
    0.06
     Erot
    0.06
    popup
    0.06
     Druh
    0.06
     sut
    0.06
     leaning
    0.06
    Activations Density 0.086%

    No Known Activations