Isabelle Cudlitz
commited on
Commit
·
8d7525b
1
Parent(s):
11ff4e2
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -2520,4 +2520,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 2520 |
### Framework versions
|
| 2521 |
|
| 2522 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2523 |
- PEFT 0.6.2
|
|
|
|
| 2520 |
### Framework versions
|
| 2521 |
|
| 2522 |
|
| 2523 |
+
- PEFT 0.6.2
|
| 2524 |
+
## Training procedure
|
| 2525 |
+
|
| 2526 |
+
|
| 2527 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 2528 |
+
- load_in_8bit: True
|
| 2529 |
+
- load_in_4bit: False
|
| 2530 |
+
- llm_int8_threshold: 6.0
|
| 2531 |
+
- llm_int8_skip_modules: None
|
| 2532 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 2533 |
+
- llm_int8_has_fp16_weight: False
|
| 2534 |
+
- bnb_4bit_quant_type: fp4
|
| 2535 |
+
- bnb_4bit_use_double_quant: False
|
| 2536 |
+
- bnb_4bit_compute_dtype: float32
|
| 2537 |
+
|
| 2538 |
+
### Framework versions
|
| 2539 |
+
|
| 2540 |
+
|
| 2541 |
- 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:c46d9dcfbf68bf3537ea56f405ba2d474969a580541bf3a30ddb474eaafc0b65
|
| 3 |
size 21020682
|