© 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. 90668
    Prev
    Next
    INDEX
    Explanations

    finance or software projects

    np_max-act · gemini-2.0-flash

    The neuron is tuned to spot multi‐syllable proper names—i.e. named entities such as person names (Rajaratnam), organization names (Galleon), and domain names (Tigris.org).

    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
    .ch
    -0.06
     babes
    -0.06
    _crit
    -0.06
    .Pod
    -0.06
    -consuming
    -0.06
    ViewPager
    -0.06
    we
    -0.06
    Chem
    -0.06
     DEM
    -0.06
    Few
    -0.06
    POSITIVE LOGITS
     dư�
    0.08
    ident
    0.07
     되는
    0.06
    ้าส
    0.06
     �
    0.06
    .program
    0.06
     mềm
    0.06
    ğını
    0.06
    	buffer
    0.06
    ffiti
    0.06
    Activations Density 0.000%

    No Known Activations