3 ms·
they stated it in the original llama paper
by perfmode 2y ago
they stated it in the original llama paper
- bredren 2y agoThe 2/27/23 paper? Where?
- rafram 2y agoRight, the original LLaMA paper. LLaMA 2 and 3 are significantly more capable models trained on orders of magnitude more data, and those papers notably do not say where the data comes from. The LLaMA 3 paper helpfully mentions that "Much of the data we utilize is obtained from the web," so I guess that's better than nothing! LLaMA 2: https://arxiv.org/pdf/2307.09288 https://arxiv.org/pdf/2307.09288 LLaMA 3: https://arxiv.org/pdf/2407.21783 https://arxiv.org/pdf/2407.21783