Isabelle Cudlitz
commited on
Commit
·
c5ccdae
1
Parent(s):
30cbcbc
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -1692,4 +1692,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 1692 |
### Framework versions
|
| 1693 |
|
| 1694 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1695 |
- PEFT 0.6.2
|
|
|
|
| 1692 |
### Framework versions
|
| 1693 |
|
| 1694 |
|
| 1695 |
+
- PEFT 0.6.2
|
| 1696 |
+
## Training procedure
|
| 1697 |
+
|
| 1698 |
+
|
| 1699 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 1700 |
+
- load_in_8bit: True
|
| 1701 |
+
- load_in_4bit: False
|
| 1702 |
+
- llm_int8_threshold: 6.0
|
| 1703 |
+
- llm_int8_skip_modules: None
|
| 1704 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 1705 |
+
- llm_int8_has_fp16_weight: False
|
| 1706 |
+
- bnb_4bit_quant_type: fp4
|
| 1707 |
+
- bnb_4bit_use_double_quant: False
|
| 1708 |
+
- bnb_4bit_compute_dtype: float32
|
| 1709 |
+
|
| 1710 |
+
### Framework versions
|
| 1711 |
+
|
| 1712 |
+
|
| 1713 |
- 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:b2a0d87ddcbf8a1e6d4236f61b4ce6efb90e3ac5f5525ae239e2aba493d4cb32
|
| 3 |
size 21020682
|