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. 43840
    Prev
    Next
    INDEX
    Explanations

    ischemia

    np_max-act · gemini-2.0-flash

    This neuron detects terminology related to tissue oxygen-deprivation injury and recovery (e.g., ischemia, reperfusion, hypoxia).

    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
     fclose
    -0.08
    fclose
    -0.07
     wastewater
    -0.07
     αλλά
    -0.07
    acias
    -0.06
    "])↵
    -0.06
     setbacks
    -0.06
    Mapper
    -0.06
    .Close
    -0.06
    	side
    -0.06
    POSITIVE LOGITS
    aligned
    0.08
     agricult
    0.07
    draul
    0.07
    .ny
    0.06
    {o
    0.06
    vertiser
    0.06
     geopol
    0.06
    料
    0.06
    periment
    0.06
     indeb
    0.06
    Activations Density 0.018%

    No Known Activations