© 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. 73622
    Prev
    Next
    INDEX
    Explanations

    attering

    np_max-act · gemini-2.0-flash

    The neuron fires on bits of dramatic, hyperbolic adjectives—especially the pieces of sensational, often hyphenated modifiers (e.g. “earth-shattering,” “hi-tech,” “super-scientific”).

    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
     Jack
    -0.06
    Cost
    -0.06
    Jack
    -0.06
    alim
    -0.06
     wer
    -0.06
     IM
    -0.06
    igel
    -0.06
    AAC
    -0.06
     cert
    -0.06
    angel
    -0.05
    POSITIVE LOGITS
     crappy
    0.07
     اخبار
    0.07
     prestigious
    0.07
    _band
    0.07
     ніч
    0.07
     العم
    0.07
    .debugLine
    0.07
    会议
    0.06
     Neon
    0.06
     vlak
    0.06
    Activations Density 0.078%

    No Known Activations