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. 92967
    Prev
    Next
    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash

    The neuron detects subtitle time‐code markers—i.e. numeric timestamps with colons and decimal fractions.

    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
    .LOGIN
    -0.06
     Girlfriend
    -0.06
    <<<
    -0.06
     JT
    -0.06
    ENG
    -0.06
    .JFrame
    -0.06
    山市
    -0.06
    �
    -0.05
     �
    -0.05
    .ci
    -0.05
    POSITIVE LOGITS
    шего
    0.08
     Бел
    0.07
    (Object
    0.07
    -default
    0.07
     nổi
    0.07
    (components
    0.07
     nuestro
    0.07
     Igor
    0.07
     verifying
    0.06
    (sum
    0.06
    Activations Density 0.001%

    No Known Activations