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. 55004
    Prev
    Next
    INDEX
    Explanations

    Sports game recaps

    np_max-act · gemini-2.0-flash

    The neuron detects mentions of scoring runs or lead changes, specifically segments describing a team’s numeric “run” (e.g. “14-5 run”) or jumping to a particular point‐differential lead.

    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
     Pest
    -0.07
     Macro
    -0.07
     Christoph
    -0.06
    utom
    -0.06
    wild
    -0.06
    _HINT
    -0.06
     Tut
    -0.06
     objectives
    -0.06
     değiş
    -0.06
     pageNumber
    -0.06
    POSITIVE LOGITS
    Sequential
    0.08
     може
    0.07
     galaxy
    0.07
     naše
    0.06
     А
    0.06
     forge
    0.06
    .nombre
    0.06
     �
    0.06
    �
    0.06
    $$
    0.06
    Activations Density 0.012%

    No Known Activations