K2: a fully-reproducible large language model outperforming Llama 2 70B using 35% less compute
LLM360 demystifies the training recipe used for Llama 2 70B with K2. K2 is fully transparent, meaning we’ve open-sourced all artifacts, including code, data, model checkpoints, intermediate results, and more.
LLM360 Model Performance and Evaluation Collection
The LLM360 Performance and Evaluation Collection is a robust evaluations set consisting of general and domain specific evaluations to assess model knowledge and function.
Evaluations include standard best practice benchmarks, medical, math, and coding knowledge. More about the evaluations can be found here.
k2 big eval table
Detailed analysis can be found on the K2 Weights and Biases project here
K2 Gallery
The K2 gallery allows one to browse the output of various prompts on intermediate K2 checkpoints, which provides an intuitive understanding on how the model develops and improves over time. This is inspired by The Bloom Book.
We provide step-by-step reproducation tutorials for tech enthusiasts, AI practitioners and academic or industry researchers who want to learn pretraining techniques here.
LLM360 Developer Suite
We provide step-by-step finetuning tutorials for tech enthusiasts, AI practitioners and academic or industry researchers here.
Loading K2
python
1from transformers import AutoModelForCausalLM, AutoTokenizer
23tokenizer = AutoTokenizer.from_pretrained("LLM360/K2")4model = AutoModelForCausalLM.from_pretrained("LLM360/K2")56prompt ='what is the highest mountain on earth?'78input_ids = tokenizer(prompt, return_tensors="pt").input_ids
9gen_tokens = model.generate(input_ids, do_sample=True, max_new_tokens=128)1011print("-"*20+"Output for model"+20*'-')12print(tokenizer.batch_decode(gen_tokens)[0])
About LLM360
LLM360 is an open research lab enabling community-owned AGI through open-source large model research and development.
LLM360 enables community-owned AGI by creating standards and tools to advance the bleeding edge of LLM capability and empower knowledge transfer, research, and development.
We believe in a future where artificial general intelligence (AGI) is created by the community, for the community. Through an open ecosystem of equitable computational resources, high quality data, and flowing technical knowledge, we can ensure ethical AGI development and universal access for all innovators.
