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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 73243
    Prev
    Next
    INDEX
    Explanations

    discourse-structuring cues and transition phrases that signal enumeration, explanation, or emphasis, especially at the beginnings of sentences.

    oai_token-act-pair · gpt-5Triggered by @vetterc0

    discourse-structuring transitions and framing phrases that organize explanations or arguments (e.g., sentence-initial markers like First, Second, Instead, In these ways, This, There).

    oai_token-act-pair · gpt-5Triggered 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
    基金经理
    -0.08
    ผล
    -0.08
     Southampton
    -0.07
    底盘
    -0.07
     קופ
    -0.07
     soy
    -0.07
    蓍
    -0.07
     ViewHolder
    -0.07
    大专
    -0.07
     headlines
    -0.07
    POSITIVE LOGITS
    Registers
    0.08
     MSR
    0.07
    露
    0.07
     CRA
    0.07
    Because
    0.07
    з
    0.07
    reviews
    0.07
     Self
    0.07
    .Region
    0.07
     Disease
    0.07
    Activations Density 0.022%

    No Known Activations