Requantize ggml after breaking updates (again)

Requantize ggml after breaking updates

Add pytorch model files

set use_cache=true in config.json

Update readme

Delete epoch3 models from this branch. Add correct configs

Add GPTQ .safetensors for epoch6

Add 4k-epoch6

set use_cache=true in config.json

Add tokenizer and configs

Add epoch3 model files

Create README.md

initial commit