Isabelle Cudlitz
commited on
Commit
·
b73409d
1
Parent(s):
d0b5f53
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -4086,4 +4086,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 4086 |
### Framework versions
|
| 4087 |
|
| 4088 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4089 |
- PEFT 0.6.2
|
|
|
|
| 4086 |
### Framework versions
|
| 4087 |
|
| 4088 |
|
| 4089 |
+
- PEFT 0.6.2
|
| 4090 |
+
## Training procedure
|
| 4091 |
+
|
| 4092 |
+
|
| 4093 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 4094 |
+
- load_in_8bit: True
|
| 4095 |
+
- load_in_4bit: False
|
| 4096 |
+
- llm_int8_threshold: 6.0
|
| 4097 |
+
- llm_int8_skip_modules: None
|
| 4098 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 4099 |
+
- llm_int8_has_fp16_weight: False
|
| 4100 |
+
- bnb_4bit_quant_type: fp4
|
| 4101 |
+
- bnb_4bit_use_double_quant: False
|
| 4102 |
+
- bnb_4bit_compute_dtype: float32
|
| 4103 |
+
|
| 4104 |
+
### Framework versions
|
| 4105 |
+
|
| 4106 |
+
|
| 4107 |
- 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:ddd45006ac74b62397c8aae038d0c7542a50f1387a4784d38127f9bcd7e31e08
|
| 3 |
size 21020682
|