© 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. 79560
    Prev
    Next
    INDEX
    Explanations

    answering questions

    np_max-act · gemini-2.0-flash

    ideas and concepts related to artificial intelligence applications.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    the neuron detects present‐participle (–ing) verb forms, especially gerunds like “evaluating,” “analyzing,” “brewing,” “judging,” or “using.”

    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
     cancell
    -0.07
     zz
    -0.07
     Vil
    -0.07
    mana
    -0.07
    Pedido
    -0.07
    asil
    -0.06
     eine
    -0.06
     Janet
    -0.06
     Expect
    -0.06
    .coll
    -0.06
    POSITIVE LOGITS
    _SN
    0.07
    械
    0.07
    επ
    0.06
    В
    0.06
     mexico
    0.06
    yntaxException
    0.06
    ▍
    0.06
    EMENT
    0.06
    '%(
    0.06
     میشود
    0.06
    Activations Density 0.333%

    No Known Activations