© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 52266
    Prev
    Next
    INDEX
    Explanations

    technical content

    np_max-act · gemini-2.0-flash

    The neuron flags long, multi‐syllable content words—especially technical or domain-specific terms (e.g. “endogenous,” “password,” “vascular,” “progressive,” “segmentation,” “rotary,” “relativity,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    technical terms and concepts related to chemical processes and materials.

    oai_token-act-pair · gpt-4o-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
     raw
    -0.07
     ordinarily
    -0.06
     ALS
    -0.06
     Bent
    -0.06
     فرم
    -0.06
    `)
    -0.06
     Serge
    -0.06
    Boost
    -0.06
     pě
    -0.06
    (PC
    -0.06
    POSITIVE LOGITS
    HasMaxLength
    0.07
    quiries
    0.07
     Netanyahu
    0.06
     interact
    0.06
    .infinity
    0.06
     KeyboardInterrupt
    0.06
    .createComponent
    0.06
    (todo
    0.06
     жовтня
    0.06
    nid
    0.06
    Activations Density 0.362%

    No Known Activations