Isabelle Cudlitz
commited on
Commit
·
4168eb5
1
Parent(s):
9461204
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -396,4 +396,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 396 |
### Framework versions
|
| 397 |
|
| 398 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 399 |
- PEFT 0.6.2
|
|
|
|
| 396 |
### Framework versions
|
| 397 |
|
| 398 |
|
| 399 |
+
- PEFT 0.6.2
|
| 400 |
+
## Training procedure
|
| 401 |
+
|
| 402 |
+
|
| 403 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 404 |
+
- load_in_8bit: False
|
| 405 |
+
- load_in_4bit: True
|
| 406 |
+
- llm_int8_threshold: 6.0
|
| 407 |
+
- llm_int8_skip_modules: None
|
| 408 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 409 |
+
- llm_int8_has_fp16_weight: False
|
| 410 |
+
- bnb_4bit_quant_type: fp4
|
| 411 |
+
- bnb_4bit_use_double_quant: True
|
| 412 |
+
- bnb_4bit_compute_dtype: float32
|
| 413 |
+
|
| 414 |
+
### Framework versions
|
| 415 |
+
|
| 416 |
+
|
| 417 |
- 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 sha256:3f457edfd1d0808213e62ba2e5810bfd63f7eaf22ac07dc8a7941bdbcc78fd4b
|
| 3 |
size 21020682
|