© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 125058
    Prev
    Next
    INDEX
    Explanations

    music reviews

    np_max-act · gemini-2.0-flash

    The neuron mainly picks out numeric metadata values—especially floating‐point numbers and measurements (e.g. durations, dates, genre weights) in the document.

    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
    记录
    -0.07
    こん
    -0.06
    課
    -0.06
    .glide
    -0.06
    .hours
    -0.06
     KBS
    -0.06
    placing
    -0.06
    +'.
    -0.06
     Accept
    -0.06
     reverse
    -0.06
    POSITIVE LOGITS
     vital
    0.07
    jev
    0.07
    Isn
    0.06
    ressing
    0.06
     vict
    0.06
     भग
    0.06
    ewire
    0.06
     chatte
    0.06
     århus
    0.06
     Abb
    0.06
    Activations Density 0.026%

    No Known Activations