LLMs which have relatively flat distribution of tokens, see https://huggingface.co/blog/ChuckMcSneed/name-diversity-in-llms-experiment
-
huggyllama/llama-65b
Text Generation β’ 65B β’ Updated β’ 1.28k β’ 77 -
meta-llama/Llama-2-70b-hf
Text Generation β’ 69B β’ Updated β’ 22.5k β’ 855 -
meta-llama/Meta-Llama-3-70B
Text Generation β’ 71B β’ Updated β’ 122k β’ β’ 875 -
meta-llama/Llama-3.1-70B
Text Generation β’ 71B β’ Updated β’ 96.6k β’ β’ 418