© 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. 43396
    Prev
    Next
    INDEX
    Explanations

    anterior

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of surgical approach terminology, especially anterior/posterior approaches and associated procedure names (e.g., decompression, fusion, corpectomy).

    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
    _complex
    -0.07
    wer
    -0.06
     switch
    -0.06
     Trigger
    -0.06
    .begin
    -0.06
    992
    -0.06
     riff
    -0.06
     différent
    -0.06
    exception
    -0.06
     Ul
    -0.06
    POSITIVE LOGITS
     recursos
    0.08
    लत
    0.07
     verbally
    0.07
     görüntü
    0.07
     ủy
    0.07
     Cognitive
    0.07
     south
    0.06
    sav
    0.06
    dateTime
    0.06
     péče
    0.06
    Activations Density 0.002%

    No Known Activations