© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 92165
    Prev
    Next
    INDEX
    Explanations

    which

    np_max-act · gemini-2.0-flash

    This neuron detects comparative/conditional phrases of the form “depending on which … you …” or “depends on which … you ….”

    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
    erm
    -0.07
    286
    -0.06
     Lis
    -0.06
    )",
↵
    -0.06
     scrambled
    -0.06
     Ко
    -0.06
    _locator
    -0.06
    Invocation
    -0.06
    sat
    -0.06
     cur
    -0.06
    POSITIVE LOGITS
    \":{\"
    0.06
     *_
    0.06
    泥
    0.06
     transferring
    0.06
     NON
    0.06
     porrf
    0.06
     inaugur
    0.06
     Gener
    0.06
     dön
    0.06
    .contentView
    0.06
    Activations Density 0.035%

    No Known Activations