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. 1231
    Prev
    Next
    INDEX
    Explanations

    shared

    np_max-act · gemini-2.0-flash

    This neuron activates on the token “shared,” i.e. it detects occurrences of the word “shared.”

    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
     Ty
    -0.08
     Bison
    -0.07
     Timothy
    -0.07
     Lion
    -0.07
     Bulletin
    -0.07
    aptor
    -0.07
    cription
    -0.07
     специ
    -0.07
    Ty
    -0.06
     cortisol
    -0.06
    POSITIVE LOGITS
     연
    0.07
    rd
    0.07
    emode
    0.06
    nd
    0.06
     kami
    0.06
    .shared
    0.06
     Shard
    0.06
    共
    0.06
     portfolio
    0.06
     торгів
    0.06
    Activations Density 0.004%

    No Known Activations