Isabelle Cudlitz
commited on
Commit
·
b747952
1
Parent(s):
c6d73e8
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -558,4 +558,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 558 |
### Framework versions
|
| 559 |
|
| 560 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 561 |
- PEFT 0.6.2
|
|
|
|
| 558 |
### Framework versions
|
| 559 |
|
| 560 |
|
| 561 |
+
- PEFT 0.6.2
|
| 562 |
+
## Training procedure
|
| 563 |
+
|
| 564 |
+
|
| 565 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 566 |
+
- load_in_8bit: True
|
| 567 |
+
- load_in_4bit: False
|
| 568 |
+
- llm_int8_threshold: 6.0
|
| 569 |
+
- llm_int8_skip_modules: None
|
| 570 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 571 |
+
- llm_int8_has_fp16_weight: False
|
| 572 |
+
- bnb_4bit_quant_type: fp4
|
| 573 |
+
- bnb_4bit_use_double_quant: False
|
| 574 |
+
- bnb_4bit_compute_dtype: float32
|
| 575 |
+
|
| 576 |
+
### Framework versions
|
| 577 |
+
|
| 578 |
+
|
| 579 |
- 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:ddb906d3004cb9fc4063bcd8352713d6e26b65b17be5bcb4f8868cfcfed899ee
|
| 3 |
size 21020682
|