© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 30855
    Prev
    Next
    INDEX
    Explanations

    it

    np_max-act · gemini-2.0-flash

    This neuron detects modal or conditional auxiliary verbs (e.g., “would,” “could,” “should”) indicating hypothetical or “would-have” constructions.

    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
     furnish
    -0.07
    462
    -0.07
    삼
    -0.07
     pager
    -0.07
     positive
    -0.07
     indication
    -0.06
     Food
    -0.06
    Services
    -0.06
     faculty
    -0.06
    یه
    -0.06
    POSITIVE LOGITS
    .win
    0.07
    -insert
    0.07
    _rt
    0.06
    oples
    0.06
    adresse
    0.06
    eview
    0.06
     exact
    0.06
     Anders
    0.05
    rire
    0.05
     Mayıs
    0.05
    Activations Density 0.063%

    No Known Activations