Update README.md
Browse files
README.md
CHANGED
|
@@ -56,7 +56,7 @@ Below is an instruction that describes a task. Write a response that appropriate
|
|
| 56 |
`nmitchko/ML1-34b-previews` is a large language model repository of LoRA checkpoints specifically fine-tuned to add text-book synthesized data in the style of Phi 1/1.5.
|
| 57 |
It is based on [`codellama-34b-hf`](https://huggingface.co/codellama/CodeLlama-34b-hf) at 34 billion parameters.
|
| 58 |
|
| 59 |
-
The primary goal of this model is to
|
| 60 |
It was trained using [LoRA](https://arxiv.org/abs/2106.09685), specifically [QLora Multi GPU](https://github.com/ChrisHayduk/qlora-multi-gpu), to reduce memory footprint.
|
| 61 |
|
| 62 |
See Training Parameters for more info This Lora supports 4-bit and 8-bit modes.
|
|
|
|
| 56 |
`nmitchko/ML1-34b-previews` is a large language model repository of LoRA checkpoints specifically fine-tuned to add text-book synthesized data in the style of Phi 1/1.5.
|
| 57 |
It is based on [`codellama-34b-hf`](https://huggingface.co/codellama/CodeLlama-34b-hf) at 34 billion parameters.
|
| 58 |
|
| 59 |
+
The primary goal of this model is to test various fine tuning methods around high quality data.
|
| 60 |
It was trained using [LoRA](https://arxiv.org/abs/2106.09685), specifically [QLora Multi GPU](https://github.com/ChrisHayduk/qlora-multi-gpu), to reduce memory footprint.
|
| 61 |
|
| 62 |
See Training Parameters for more info This Lora supports 4-bit and 8-bit modes.
|