Commit History
revert(config): use `float16` torch dtype
9ed3f94
verified
Update README.md
e4652cb
verified
Fix base model link (#11)
28b4cfb
verified
fix(modeling): use correct `base_model_prefix` name
fa88b77
verified
OpenVINO NNCF 4BIT quantization
9e73e07
Ashish
commited on
fix(tokenizer): expose `errors`
e795a4e
verified
GGUF Q4_0, Q4_1, Q8_0 quantized files
8b1a48d
ashishdatta
commited on
feat: add dropout support
8e5b1aa
fix: make `eos_token`/`pad_token` overridable and add `pickle` support
589adbf
verified
GGUF Q5_K_M quantize
4ae0672
ashishdatta
commited on
FP16 GGUF file
6d37092
ashishdatta
commited on