Isabelle Cudlitz
commited on
Commit
·
3b05f1e
1
Parent(s):
27cdd04
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -1602,4 +1602,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 1602 |
### Framework versions
|
| 1603 |
|
| 1604 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1605 |
- PEFT 0.6.2
|
|
|
|
| 1602 |
### Framework versions
|
| 1603 |
|
| 1604 |
|
| 1605 |
+
- PEFT 0.6.2
|
| 1606 |
+
## Training procedure
|
| 1607 |
+
|
| 1608 |
+
|
| 1609 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 1610 |
+
- load_in_8bit: True
|
| 1611 |
+
- load_in_4bit: False
|
| 1612 |
+
- llm_int8_threshold: 6.0
|
| 1613 |
+
- llm_int8_skip_modules: None
|
| 1614 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 1615 |
+
- llm_int8_has_fp16_weight: False
|
| 1616 |
+
- bnb_4bit_quant_type: fp4
|
| 1617 |
+
- bnb_4bit_use_double_quant: False
|
| 1618 |
+
- bnb_4bit_compute_dtype: float32
|
| 1619 |
+
|
| 1620 |
+
### Framework versions
|
| 1621 |
+
|
| 1622 |
+
|
| 1623 |
- 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:ef2de7509b1694aa83ddb58cc6bcea21314699ae7e07b54c6e9b886c6331411a
|
| 3 |
size 21020682
|