© 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. Llama3.3-70B-IT
    5. 50-RESID-POST-GF
    7. 33516
    Prev
    Next
    INDEX
    Explanations

    AI and human comparison

    np_acts-logits-general · gemini-2.5-flash-lite

    statements that label or explain a post as automated/bot-generated, often with meta-descriptive phrasing and transitional cues like sentence-initial connectors after a line break.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.3-70B-IT @ 50-resid-post-gf
    Configuration
    Goodfire/Llama-3.3-70B-Instruct-SAE-l50/Llama-3.3-70B-Instruct-SAE-l50.pt
    Prompts (Dashboard)
    10,000 prompts, 128 tokens each
    Dataset (Dashboard)
    lmsys/lmsys-chat-1m
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ãģħ
    -0.10
    Editable
    -0.10
    nels
    -0.09
     ï¾ĺ
    -0.09
    ego
    -0.09
     Rug
    -0.09
    ivan
    -0.09
     Portable
    -0.08
     Wireless
    -0.08
     Bodies
    -0.08
    POSITIVE LOGITS
     machines
    0.48
     machine
    0.46
     Machine
    0.34
     Machines
    0.33
    machine
    0.33
     AI
    0.31
    Machine
    0.30
    -machine
    0.30
    æľº
    0.29
    æ©Ł
    0.28
    Activations Density 0.594%

    No Known Activations