Isabelle Cudlitz
commited on
Commit
·
82231c5
1
Parent(s):
f5623b1
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -1386,4 +1386,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 1386 |
### Framework versions
|
| 1387 |
|
| 1388 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1389 |
- PEFT 0.6.2
|
|
|
|
| 1386 |
### Framework versions
|
| 1387 |
|
| 1388 |
|
| 1389 |
+
- PEFT 0.6.2
|
| 1390 |
+
## Training procedure
|
| 1391 |
+
|
| 1392 |
+
|
| 1393 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 1394 |
+
- load_in_8bit: False
|
| 1395 |
+
- load_in_4bit: True
|
| 1396 |
+
- llm_int8_threshold: 6.0
|
| 1397 |
+
- llm_int8_skip_modules: None
|
| 1398 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 1399 |
+
- llm_int8_has_fp16_weight: False
|
| 1400 |
+
- bnb_4bit_quant_type: fp4
|
| 1401 |
+
- bnb_4bit_use_double_quant: True
|
| 1402 |
+
- bnb_4bit_compute_dtype: float32
|
| 1403 |
+
|
| 1404 |
+
### Framework versions
|
| 1405 |
+
|
| 1406 |
+
|
| 1407 |
- 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 sha256:0c68176d60c66f48b3567b289d44110e6fac34825eed90829a97cf3725e20c2a
|
| 3 |
size 21020682
|