Isabelle Cudlitz
commited on
Commit
·
e0ded0f
1
Parent(s):
6af0e43
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -3078,4 +3078,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 3078 |
### Framework versions
|
| 3079 |
|
| 3080 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3081 |
- PEFT 0.6.2
|
|
|
|
| 3078 |
### Framework versions
|
| 3079 |
|
| 3080 |
|
| 3081 |
+
- PEFT 0.6.2
|
| 3082 |
+
## Training procedure
|
| 3083 |
+
|
| 3084 |
+
|
| 3085 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 3086 |
+
- load_in_8bit: True
|
| 3087 |
+
- load_in_4bit: False
|
| 3088 |
+
- llm_int8_threshold: 6.0
|
| 3089 |
+
- llm_int8_skip_modules: None
|
| 3090 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 3091 |
+
- llm_int8_has_fp16_weight: False
|
| 3092 |
+
- bnb_4bit_quant_type: fp4
|
| 3093 |
+
- bnb_4bit_use_double_quant: False
|
| 3094 |
+
- bnb_4bit_compute_dtype: float32
|
| 3095 |
+
|
| 3096 |
+
### Framework versions
|
| 3097 |
+
|
| 3098 |
+
|
| 3099 |
- 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:ccbb00ec1440b6c4c3bcc0ab35b1972fe5bd8283ff2e842740f0bf000d0973fe
|
| 3 |
size 21020682
|