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. 86100
    Prev
    Next
    INDEX
    Explanations

    marijuana businesses

    np_max-act · gemini-2.0-flash

    regulatory terms and their implications regarding cannabis legislation.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron detects numeric tokens representing decimal‐style measurements or values (e.g., floating‐point numbers).

    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
     tree
    -0.09
    865
    -0.08
    -web
    -0.07
    self
    -0.07
     leaf
    -0.07
     sdl
    -0.07
     amplifier
    -0.07
     Sponge
    -0.06
     cluster
    -0.06
    tree
    -0.06
    POSITIVE LOGITS
    ulos
    0.06
    .Stdout
    0.06
     vulner
    0.06
    =$((
    0.06
     bahwa
    0.06
    规定
    0.06
    ]={↵
    0.05
    [current
    0.05
     ナ
    0.05
     khảo
    0.05
    Activations Density 0.012%

    No Known Activations