Isabelle Cudlitz
commited on
Commit
·
03487d8
1
Parent(s):
09588c9
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -2196,4 +2196,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 2196 |
### Framework versions
|
| 2197 |
|
| 2198 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2199 |
- PEFT 0.6.2
|
|
|
|
| 2196 |
### Framework versions
|
| 2197 |
|
| 2198 |
|
| 2199 |
+
- PEFT 0.6.2
|
| 2200 |
+
## Training procedure
|
| 2201 |
+
|
| 2202 |
+
|
| 2203 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 2204 |
+
- load_in_8bit: True
|
| 2205 |
+
- load_in_4bit: False
|
| 2206 |
+
- llm_int8_threshold: 6.0
|
| 2207 |
+
- llm_int8_skip_modules: None
|
| 2208 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 2209 |
+
- llm_int8_has_fp16_weight: False
|
| 2210 |
+
- bnb_4bit_quant_type: fp4
|
| 2211 |
+
- bnb_4bit_use_double_quant: False
|
| 2212 |
+
- bnb_4bit_compute_dtype: float32
|
| 2213 |
+
|
| 2214 |
+
### Framework versions
|
| 2215 |
+
|
| 2216 |
+
|
| 2217 |
- 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 sha256:24b2f5e231e3a26df3ec2805ec8e3095cadaad3a8c21a02062659de057148038
|
| 3 |
size 21020682
|