Isabelle Cudlitz
commited on
Commit
·
8ed01d2
1
Parent(s):
eb8f03b
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -3240,4 +3240,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 3240 |
### Framework versions
|
| 3241 |
|
| 3242 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3243 |
- PEFT 0.6.2
|
|
|
|
| 3240 |
### Framework versions
|
| 3241 |
|
| 3242 |
|
| 3243 |
+
- PEFT 0.6.2
|
| 3244 |
+
## Training procedure
|
| 3245 |
+
|
| 3246 |
+
|
| 3247 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 3248 |
+
- load_in_8bit: True
|
| 3249 |
+
- load_in_4bit: False
|
| 3250 |
+
- llm_int8_threshold: 6.0
|
| 3251 |
+
- llm_int8_skip_modules: None
|
| 3252 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 3253 |
+
- llm_int8_has_fp16_weight: False
|
| 3254 |
+
- bnb_4bit_quant_type: fp4
|
| 3255 |
+
- bnb_4bit_use_double_quant: False
|
| 3256 |
+
- bnb_4bit_compute_dtype: float32
|
| 3257 |
+
|
| 3258 |
+
### Framework versions
|
| 3259 |
+
|
| 3260 |
+
|
| 3261 |
- 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:fe69d9addb9de4b6c4413c5f0434de9bdd6aa6c11cafcd3cdd504cafc87c8435
|
| 3 |
size 21020682
|