© 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
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 68775
    Prev
    Next
    INDEX
    Explanations

    can

    np_max-act · gemini-2.0-flash

    This neuron fires on first-person self-references (words like “I,” “my,” “can,” “access,” etc.) indicating the author’s own intent or actions.

    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
    _lit
    -0.07
    Spin
    -0.07
    98
    -0.07
    리아
    -0.06
    DownLatch
    -0.06
    .song
    -0.06
    anes
    -0.06
    ــــ
    -0.06
     pedestal
    -0.06
    周收录
    -0.06
    POSITIVE LOGITS
     autopsy
    0.06
    jít
    0.06
    .struts
    0.06
    ौद
    0.06
    خصوص
    0.06
     framed
    0.06
    <Field
    0.06
     rehab
    0.06
     reins
    0.06
     pne
    0.06
    Activations Density 0.034%

    No Known Activations