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. 43726
    Prev
    Next
    INDEX
    Explanations

    sports scores

    np_max-act · gemini-2.0-flash

    This neuron detects numeric scorelines (e.g. “3-1”, “7-0”, “2-1”) in sports reporting.

    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
     someday
    -0.06
    世
    -0.06
     preds
    -0.06
    study
    -0.06
     Wire
    -0.06
     Cabinets
    -0.06
    scopy
    -0.06
    -paper
    -0.06
    pas
    -0.06
    Transfer
    -0.05
    POSITIVE LOGITS
     �
    0.07
     SUBJECT
    0.07
     [|
    0.07
    .setItems
    0.06
     Decomp
    0.06
     Nicole
    0.06
     buc
    0.06
    ?):
    0.06
     reinforcements
    0.06
     qry
    0.06
    Activations Density 0.007%

    No Known Activations