Isabelle Cudlitz
commited on
Commit
·
4448e68
1
Parent(s):
84714af
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -2556,4 +2556,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 2556 |
### Framework versions
|
| 2557 |
|
| 2558 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2559 |
- PEFT 0.6.2
|
|
|
|
| 2556 |
### Framework versions
|
| 2557 |
|
| 2558 |
|
| 2559 |
+
- PEFT 0.6.2
|
| 2560 |
+
## Training procedure
|
| 2561 |
+
|
| 2562 |
+
|
| 2563 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 2564 |
+
- load_in_8bit: True
|
| 2565 |
+
- load_in_4bit: False
|
| 2566 |
+
- llm_int8_threshold: 6.0
|
| 2567 |
+
- llm_int8_skip_modules: None
|
| 2568 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 2569 |
+
- llm_int8_has_fp16_weight: False
|
| 2570 |
+
- bnb_4bit_quant_type: fp4
|
| 2571 |
+
- bnb_4bit_use_double_quant: False
|
| 2572 |
+
- bnb_4bit_compute_dtype: float32
|
| 2573 |
+
|
| 2574 |
+
### Framework versions
|
| 2575 |
+
|
| 2576 |
+
|
| 2577 |
- 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 sha256:79bffe713755002c4a0e31cd6754218d0ef25adf96651ab2f662013d6ca20fec
|
| 3 |
size 21020682
|