Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 47
    Prev
    Next
    INDEX
    Explanations

    freedom and liberation

    np_max-act · gemini-2.0-flash

    The neuron responds to political-history terms—especially words tied to independence or liberation (e.g. “freedom,” “liberation,” “independence,” “Risorgimento,” “struggle”) and the names of prominent historical leaders.

    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
    Short
    -0.06
    zion
    -0.06
    edriver
    -0.06
     animate
    -0.06
     ruins
    -0.06
    portlet
    -0.06
    nak
    -0.06
     Vital
    -0.06
     �
    -0.06
     forks
    -0.06
    POSITIVE LOGITS
     Henri
    0.07
     happier
    0.07
    tej
    0.06
    chosen
    0.06
     consulted
    0.06
     kez
    0.06
    (mouse
    0.06
    テ
    0.06
     Expanded
    0.06
     mocked
    0.06
    Activations Density 0.043%

    No Known Activations