Isabelle Cudlitz
commited on
Commit
·
ca040c8
1
Parent(s):
54a6669
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -522,4 +522,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 522 |
### Framework versions
|
| 523 |
|
| 524 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 525 |
- PEFT 0.6.2
|
|
|
|
| 522 |
### Framework versions
|
| 523 |
|
| 524 |
|
| 525 |
+
- PEFT 0.6.2
|
| 526 |
+
## Training procedure
|
| 527 |
+
|
| 528 |
+
|
| 529 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 530 |
+
- load_in_8bit: False
|
| 531 |
+
- load_in_4bit: True
|
| 532 |
+
- llm_int8_threshold: 6.0
|
| 533 |
+
- llm_int8_skip_modules: None
|
| 534 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 535 |
+
- llm_int8_has_fp16_weight: False
|
| 536 |
+
- bnb_4bit_quant_type: fp4
|
| 537 |
+
- bnb_4bit_use_double_quant: True
|
| 538 |
+
- bnb_4bit_compute_dtype: float32
|
| 539 |
+
|
| 540 |
+
### Framework versions
|
| 541 |
+
|
| 542 |
+
|
| 543 |
- 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 sha256:93e67182c4e03274d129b0a9aa43f6163fdcb2446ff8599bd8b7e1220603e46e
|
| 3 |
size 21020682
|