© 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
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 7876
    Prev
    Next
    INDEX
    Explanations

    research, introducing

    np_max-act · gemini-2.0-flash

    This neuron fires on verbs that present or introduce the authors’ contributions or findings (e.g. “can,” “present,” “introduce,” “give”).

    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
    Coach
    -0.09
    яс
    -0.07
     بار
    -0.06
    MouseDown
    -0.06
    iliate
    -0.06
    Hat
    -0.06
     Apprentice
    -0.06
    igate
    -0.06
     Nob
    -0.06
    otor
    -0.06
    POSITIVE LOGITS
     attacking
    0.07
     Runtime
    0.07
    いの
    0.06
    िनट
    0.06
     --------
    0.06
     PyQt
    0.06
    522
    0.06
    inside
    0.06
     perfected
    0.06
     FROM
    0.06
    Activations Density 0.111%

    No Known Activations