© 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. 109729
    Prev
    Next
    INDEX
    Explanations

    agle

    np_max-act · gemini-2.0-flash

    This neuron activates on occurrences of the name “Eagle” (i.e. the token pieces forming the word “Eagle”).

    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
     Bed
    -0.07
     jsx
    -0.07
    (scene
    -0.07
    Christopher
    -0.07
    BufferData
    -0.07
     hiatus
    -0.06
     Někter
    -0.06
     Merch
    -0.06
    116
    -0.06
    Rem
    -0.06
    POSITIVE LOGITS
     Eagle
    0.16
     eagle
    0.12
     Eagles
    0.12
     Crow
    0.09
     Hawks
    0.09
     crow
    0.09
     Falcon
    0.09
     Falcons
    0.08
     Raven
    0.08
    Crow
    0.08
    Activations Density 0.010%

    No Known Activations