metadata
license: cc-by-sa-4.0
datasets:
- acrastt/EverythingLM-V3-ShareGPT
language:
- en
library_name: transformers
pipeline_tag: text-generation
This is StableLM 3B 4E1T finetuned on EverythingLM Data V3(ShareGPT Format) for 2 epochs using QLoRA.
Prompt template:
### HUMAN:
{prompt}
### RESPONSE:
Note that this model have the EOS token of <|endoftext|>
instead of <\s>
.
Attribution
StableLM 3B 4E1T by Stability AI is licensed under CC BY-SA 4.0.
Modifications:
- Instruction tuned on dataset EverythingLM Data V3(ShareGPT Format) for 2 epochs using QLoRA.
- Added
"_name_or_path"
, changed"torch_dtype"
and"transformers_version"
inconfig.json
. - Changes to
README.MD
.