Isabelle Cudlitz
commited on
Commit
·
983c444
1
Parent(s):
e7da36c
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -1998,4 +1998,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 1998 |
### Framework versions
|
| 1999 |
|
| 2000 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2001 |
- PEFT 0.6.2
|
|
|
|
| 1998 |
### Framework versions
|
| 1999 |
|
| 2000 |
|
| 2001 |
+
- PEFT 0.6.2
|
| 2002 |
+
## Training procedure
|
| 2003 |
+
|
| 2004 |
+
|
| 2005 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 2006 |
+
- load_in_8bit: True
|
| 2007 |
+
- load_in_4bit: False
|
| 2008 |
+
- llm_int8_threshold: 6.0
|
| 2009 |
+
- llm_int8_skip_modules: None
|
| 2010 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 2011 |
+
- llm_int8_has_fp16_weight: False
|
| 2012 |
+
- bnb_4bit_quant_type: fp4
|
| 2013 |
+
- bnb_4bit_use_double_quant: False
|
| 2014 |
+
- bnb_4bit_compute_dtype: float32
|
| 2015 |
+
|
| 2016 |
+
### Framework versions
|
| 2017 |
+
|
| 2018 |
+
|
| 2019 |
- PEFT 0.6.2
|
adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 21020682
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6660c68e7943acf8b0b127e79d63b8775fa472b57be6117d1a6448253c7a2eb9
|
| 3 |
size 21020682
|