© 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. Qwen2.5-7B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 65314
    Prev
    Next
    INDEX
    Explanations

    gh

    np_max-act · gemini-2.0-flash

    The neuron detects tokens that are part of named entities, headings, or identifiers — i.e., proper nouns, titles/tags, and code/library names.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-qwen2.5-7b-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
    2
    -0.09
    3
    -0.09
    6
    -0.08
    ，
    -0.08
     Launch
    -0.07
     have
    -0.07
    _attributes
    -0.07
    想要
    -0.07
    在
    -0.07
    .Panel
    -0.07
    POSITIVE LOGITS
     underworld
    0.07
    _CARD
    0.06
     בלבד
    0.06
    0.06
    充满
    0.06
     seul
    0.06
    ظاهرة
    0.06
    ridor
    0.06
     or
    0.06
     blackout
    0.06
    Activations Density 48.312%

    No Known Activations