© 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. Qwen2.5-7B-IT
    7. Resid Post - 131k
    9. 15-RESID-POST-AA
    11. 1328
    Prev
    Next
    INDEX
    Explanations

    you I

    np_max-act · gemini-2.0-flash

    It detects assistant phrases that offer help or guidance — e.g., introductions to examples, templates, steps, or instructions.

    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_15/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ㋴
    -0.07
     śmierci
    -0.07
    ################################################################################↵
    -0.07
     love
    -0.06
    💘
    -0.06
    カテゴリ
    -0.06
    цит
    -0.06
    .major
    -0.06
    direction
    -0.06
    iedade
    -0.06
    POSITIVE LOGITS
    赳
    0.08
    .measure
    0.08
    ปราก
    0.07
    uges
    0.07
    ဘ
    0.07
     Neck
    0.07
    CanBeConverted
    0.07
    .Angle
    0.07
    쯤
    0.07
    ষ
    0.07
    Activations Density 0.138%

    No Known Activations