Commit
·
193e0b0
1
Parent(s):
93414a9
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,8 +1,12 @@
|
|
| 1 |
---
|
| 2 |
library_name: peft
|
| 3 |
---
|
| 4 |
-
|
|
|
|
|
|
|
|
|
|
| 5 |
|
|
|
|
| 6 |
|
| 7 |
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
- load_in_8bit: True
|
|
@@ -14,7 +18,5 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 14 |
- bnb_4bit_quant_type: fp4
|
| 15 |
- bnb_4bit_use_double_quant: False
|
| 16 |
- bnb_4bit_compute_dtype: float32
|
| 17 |
-
|
| 18 |
-
|
| 19 |
-
|
| 20 |
- PEFT 0.4.0
|
|
|
|
| 1 |
---
|
| 2 |
library_name: peft
|
| 3 |
---
|
| 4 |
+
# Introduction
|
| 5 |
+
Vicuna-style model based on LoRA tuning
|
| 6 |
+
- Dataset: https://huggingface.co/datasets/Aeala/ShareGPT_Vicuna_unfiltered
|
| 7 |
+
- Base model: https://huggingface.co/huggyllama/llama-7b
|
| 8 |
|
| 9 |
+
# Training procedure
|
| 10 |
|
| 11 |
The following `bitsandbytes` quantization config was used during training:
|
| 12 |
- load_in_8bit: True
|
|
|
|
| 18 |
- bnb_4bit_quant_type: fp4
|
| 19 |
- bnb_4bit_use_double_quant: False
|
| 20 |
- bnb_4bit_compute_dtype: float32
|
| 21 |
+
## Framework versions
|
|
|
|
|
|
|
| 22 |
- PEFT 0.4.0
|