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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 84575
    Prev
    Next
    INDEX
    Explanations

    remnants and residuals

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of specific anatomical features—especially the presence or “remnant” of a structure—in medical descriptive text.

    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
     uncon
    -0.06
    τά
    -0.06
     n
    -0.06
    itect
    -0.06
     imageSize
    -0.06
     найбільш
    -0.06
    translator
    -0.06
     Beckham
    -0.06
    .Toolkit
    -0.06
    543
    -0.06
    POSITIVE LOGITS
     nhóm
    0.07
    _Insert
    0.07
    (lst
    0.07
     Byl
    0.06
    Truthy
    0.06
     clown
    0.06
    ,strong
    0.06
    (By
    0.06
     Conspiracy
    0.06
    ent
    0.06
    Activations Density 0.173%

    No Known Activations