Isabelle Cudlitz
commited on
Commit
·
c26c6e8
1
Parent(s):
efa9448
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -1494,4 +1494,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 1494 |
### Framework versions
|
| 1495 |
|
| 1496 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1497 |
- PEFT 0.6.2
|
|
|
|
| 1494 |
### Framework versions
|
| 1495 |
|
| 1496 |
|
| 1497 |
+
- PEFT 0.6.2
|
| 1498 |
+
## Training procedure
|
| 1499 |
+
|
| 1500 |
+
|
| 1501 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 1502 |
+
- load_in_8bit: True
|
| 1503 |
+
- load_in_4bit: False
|
| 1504 |
+
- llm_int8_threshold: 6.0
|
| 1505 |
+
- llm_int8_skip_modules: None
|
| 1506 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 1507 |
+
- llm_int8_has_fp16_weight: False
|
| 1508 |
+
- bnb_4bit_quant_type: fp4
|
| 1509 |
+
- bnb_4bit_use_double_quant: False
|
| 1510 |
+
- bnb_4bit_compute_dtype: float32
|
| 1511 |
+
|
| 1512 |
+
### Framework versions
|
| 1513 |
+
|
| 1514 |
+
|
| 1515 |
- 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 sha256:5f237b76308cb71efc8449fddcfa1e0bc2e77aa4c632fec116742ebdb2b1ad3f
|
| 3 |
size 21020682
|