© 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. 13476
    Prev
    Next
    INDEX
    Explanations

    to

    np_max-act · gemini-2.0-flash

    The neuron activates on the token “to,” picking out occurrences of that common infinitive/preposition.

    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
    .SelectedItems
    -0.07
     fitted
    -0.06
    loadModel
    -0.06
    -metal
    -0.06
     epith
    -0.06
     manager
    -0.06
     Adrian
    -0.06
     　　　　　　　　　　　　　　　　
    -0.06
     tagged
    -0.06
     medications
    -0.06
    POSITIVE LOGITS
    \Has
    0.08
    ンス
    0.08
     ธ
    0.07
    재
    0.07
    ivor
    0.07
    좌
    0.07
    romise
    0.07
    ;]/
    0.07
    /sp
    0.07
    �
    0.07
    Activations Density 0.017%

    No Known Activations