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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 68363
    Prev
    Next
    INDEX
    Explanations

    time units

    np_max-act · gemini-2.0-flash

    discussions about the quantity and color of painted items, particularly in a hypothetical context.

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

    The neuron activates most strongly on the Chinese character “天” (day), so it detects mentions of durations in days.

    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
    tier
    -0.07
    ocs
    -0.07
    OLT
    -0.07
    .yang
    -0.06
    ože
    -0.06
    fall
    -0.06
     Intellectual
    -0.06
    _paper
    -0.06
     AH
    -0.06
    Telefono
    -0.06
    POSITIVE LOGITS
    �
    0.07
     fgets
    0.06
     supra
    0.06
    _nth
    0.06
    compileComponents
    0.06
     HRESULT
    0.06
    ubo
    0.06
     seamlessly
    0.06
    -Feb
    0.06
    .↵↵↵↵↵↵↵↵↵↵↵↵
    0.06
    Activations Density 0.006%

    No Known Activations