Small correction on number of epochs.

Looks like the model description was copied over from https://huggingface.co/nomic-ai/gpt4all-lora , so needs to be slightly tweaked.

README.md

@@ -8,10 +8,10 @@ language:
 
 # gpt4all-lora-epoch-3
 
-This is an intermediate (epoch 3 / 4) checkpoint from `nomic-ai/gpt4all-lora`.
+This is an intermediate (epoch 3 / 4) checkpoint from [`nomic-ai/gpt4all-lora`](https://huggingface.co/nomic-ai/gpt4all-lora).
 
 An autoregressive transformer trained on [data](https://huggingface.co/datasets/nomic-ai/gpt4all_prompt_generations) curated using [Atlas](https://atlas.nomic.ai/).
-This model is trained with four full epochs of training, while the related [gpt4all-lora-epoch-2 model](https://huggingface.co/nomic-ai/gpt4all-lora-epoch-2) is trained with three.
+This model is trained with three epochs of training, while the related [gpt4all-lora model](https://huggingface.co/nomic-ai/gpt4all-lora) is trained with four.
 Replication instructions and data: [https://github.com/nomic-ai/gpt4all](https://github.com/nomic-ai/gpt4all)
 
 ## Model Details