Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 130458
    Prev
    Next
    INDEX
    Explanations

    technical processes

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on technical or domain‐specific multi-syllable terms (e.g. specialized materials, processes, or equipment names).

    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
    060
    -0.08
     getProduct
    -0.07
    ์เพ
    -0.07
    atches
    -0.07
    zier
    -0.07
    095
    -0.06
    トル
    -0.06
    dato
    -0.06
     females
    -0.06
     Middle
    -0.06
    POSITIVE LOGITS
     จำนวน
    0.06
     於
    0.06
     Visualization
    0.06
    \AppData
    0.06
     поли
    0.05
    utility
    0.05
     Flake
    0.05
    MBOL
    0.05
     haired
    0.05
     framework
    0.05
    Activations Density 0.079%

    No Known Activations