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. 129033
    Prev
    Next
    INDEX
    Explanations

    stop words

    np_max-act · gemini-2.0-flash

    financial terms related to stock equity and investments.

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

    The neuron is primarily triggered by standalone decimal numbers (i.e. tokens representing floating-point values).

    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
     interfer
    -0.06
     тебя
    -0.06
     INLINE
    -0.06
    /templates
    -0.06
    .Dataset
    -0.06
     setUser
    -0.06
    EFR
    -0.06
    Word
    -0.06
    itten
    -0.06
    Logging
    -0.06
    POSITIVE LOGITS
    واع
    0.06
    habit
    0.06
    elli
    0.06
     vorhand
    0.06
     свойства
    0.06
     mục
    0.06
    ULER
    0.06
    deniz
    0.06
    CVE
    0.06
    }"↵
    0.06
    Activations Density 0.005%

    No Known Activations