Isabelle Cudlitz
commited on
Commit
·
eb8f03b
1
Parent(s):
9be15db
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -3222,4 +3222,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 3222 |
### Framework versions
|
| 3223 |
|
| 3224 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3225 |
- PEFT 0.6.2
|
|
|
|
| 3222 |
### Framework versions
|
| 3223 |
|
| 3224 |
|
| 3225 |
+
- PEFT 0.6.2
|
| 3226 |
+
## Training procedure
|
| 3227 |
+
|
| 3228 |
+
|
| 3229 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 3230 |
+
- load_in_8bit: True
|
| 3231 |
+
- load_in_4bit: False
|
| 3232 |
+
- llm_int8_threshold: 6.0
|
| 3233 |
+
- llm_int8_skip_modules: None
|
| 3234 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 3235 |
+
- llm_int8_has_fp16_weight: False
|
| 3236 |
+
- bnb_4bit_quant_type: fp4
|
| 3237 |
+
- bnb_4bit_use_double_quant: False
|
| 3238 |
+
- bnb_4bit_compute_dtype: float32
|
| 3239 |
+
|
| 3240 |
+
### Framework versions
|
| 3241 |
+
|
| 3242 |
+
|
| 3243 |
- 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:f78faffe118c46cd0e89dd048d44d451f7c7bf0d930a08a83c4a51cb7fda48ff
|
| 3 |
size 21020682
|