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. 84793
    Prev
    Next
    INDEX
    Explanations

    both

    np_max-act · gemini-2.0-flash

    This neuron detects numerical and quantitative measurement expressions (e.g. numbers, decimals, units, and comparative quantity contexts).

    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
     Kron
    -0.07
     Nẵng
    -0.07
     Khu
    -0.06
    Ltd
    -0.06
    utors
    -0.06
    flu
    -0.06
     Shank
    -0.06
    -os
    -0.06
    _stream
    -0.06
    Tree
    -0.06
    POSITIVE LOGITS
    (platform
    0.07
    acky
    0.06
    belie
    0.06
    )index
    0.06
     imprisonment
    0.06
    )|(
    0.06
    idious
    0.06
     路
    0.06
    设计器
    0.06
    decimal
    0.06
    Activations Density 0.040%

    No Known Activations