© 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. 129647
    Prev
    Next
    INDEX
    Explanations

    won

    np_max-act · gemini-2.0-flash

    The neuron specializes in spotting descriptions of a competitor winning or securing a title (e.g. “won,” “defeating,” “champion,” “beating”) in sports event summaries.

    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
     Modeling
    -0.06
    esub
    -0.06
    _TIME
    -0.06
     Pied
    -0.06
    peration
    -0.06
     proceeds
    -0.06
     planted
    -0.06
     did
    -0.06
    itzerland
    -0.06
     elic
    -0.06
    POSITIVE LOGITS
     числе
    0.07
    .setMaximum
    0.06
    另
    0.06
    '[
    0.06
    FillColor
    0.06
    �
    0.06
     derail
    0.06
    	↵↵
    0.06
     تعداد
    0.06
    озмож
    0.06
    Activations Density 0.021%

    No Known Activations