Update collated assets CSV.

resources/all_assets.csv

@@ -604,6 +604,7 @@ model,Tulu 2 DPO,AI2,"Tulu 2 DPO is created in a similar manner to Tulu 2, but w
 model,Code Tulu 2,AI2,"Code Tulu 2 is a fine-tuned version of Code LLaMA that was trained on a mix of publicly available, synthetic and human datasets.",2023-11-20,https://arxiv.org/pdf/2311.10702.pdf,https://huggingface.co/allenai/codetulu-2-13b,"text; code, text",Evaluated on MT-Bench and AlpacaEval. compared to other chatbots.,13B parameters (dense),"['Code LLaMA', 'Tulu-V2-mix']",unknown,unknown,unknown,,open,AI2 ImpACT,,,,https://huggingface.co/allenai/codetulu-2-13b/discussions,,,,,,,,,,,
 model,OLMo,AI2,"Open Language Model (OLMo) is designed to provide access to data, training code, models, and evaluation code necessary to advance AI through open research to empower academics and researchers to study the science of language models collectively.",2024-02-01,https://allenai.org/olmo/olmo-paper.pdf,https://huggingface.co/allenai/OLMo-7B,text; text,"Evaluated on standard LLM tasks and benchmarks in comparison to LLaMA, Falcon, and MPT, in addition to other same-sized models.",7B parameters (dense),['Dolma'],75.05 tCo2eq,unknown,"27 nodes, with each node containing 8x NVIDIA A100-40GB GPUs provided by MosaicML",training data from Dolma filtered and deduplicated before being trained on.,open,Apache 2.0,,,unknown,https://huggingface.co/allenai/OLMo-7B/discussions,,,,,,,,,,,
 dataset,MADLAD-400,AI2,"MADLAD-400 is a document-level multilingual dataset based on Common Crawl, covering 419 languages in total.",2023-09-09,https://arxiv.org/abs/2309.04662,,text,,3 trillion tokens,['Common Crawl'],,,,,open,CC BY 4.0,,,unknown,https://huggingface.co/datasets/allenai/MADLAD-400/discussions,,,,,,,https://huggingface.co/datasets/allenai/MADLAD-400,[],,,
+model,Evo,Liquid AI,"Evo is a foundation model based on StripedHyena, which is a deep signal processing architecture. It is capable of generalizing across DNA, RNA, and proteins and is used for generative design of new CRISPR systems.",2024-09-30,https://www.liquid.ai/research/liquid-neural-networks-research,unknown,text; text,Extensive scaling laws analysis has been performed on Evo considering beyond-transformer architectures.,unknown,['StripedHyena'],unknown,unknown,unknown,The model performance has been extensively evaluated with scaling laws analysis.,closed,unknown,"The Evo model is intended for generative design of new CRISPR systems, and it generalizes across DNA, RNA, and proteins.",unknown,unknown,Feedback or inquiries can be addressed to the Liquid AI team.,,,,,,,,,,,
 application,LinkedIn,LinkedIn,"More than 40 percent of LinkedIn's feed posts include at least one image. We want every member to have equal access to opportunity and are committed to ensuring that we make images accessible to our members who are blind or who have low vision so they can be a part of the online conversation. With Azure Cognitive Service for Vision, we can provide auto-captioning to edit and support alt. text descriptions.",,https://www.linkedin.com/,,,,,['Azure Cognitive Services for Vision'],,,,,open,unknown,,,,,,,,,,,,,,,
 application,Crisis Contact Simulator,The Trevor Project,"Crisis Contact Simulator, developed as part of a collaboration with Google.org, helps train The Trevor Project counselors by mimicking to be a teen in crisis. Crisis Contact Simulator is used as part of the training programs for the Trevor Project's 24/7 digital crisis services that supports LGBTQ youth [[Trevor Project Blog]](https://www.thetrevorproject.org/blog/the-trevor-project-launches-new-ai-tool-to-support-crisis-counselor-training/).
 ",2021-03-24,https://www.thetrevorproject.org/,,,,,['OpenAI API'],,,,Limited release,closed,unknown,"Training counselors