Isabelle Cudlitz
commited on
Commit
·
b07d8c2
1
Parent(s):
0e7c481
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -2448,4 +2448,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 2448 |
### Framework versions
|
| 2449 |
|
| 2450 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2451 |
- PEFT 0.6.2
|
|
|
|
| 2448 |
### Framework versions
|
| 2449 |
|
| 2450 |
|
| 2451 |
+
- PEFT 0.6.2
|
| 2452 |
+
## Training procedure
|
| 2453 |
+
|
| 2454 |
+
|
| 2455 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 2456 |
+
- load_in_8bit: True
|
| 2457 |
+
- load_in_4bit: False
|
| 2458 |
+
- llm_int8_threshold: 6.0
|
| 2459 |
+
- llm_int8_skip_modules: None
|
| 2460 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 2461 |
+
- llm_int8_has_fp16_weight: False
|
| 2462 |
+
- bnb_4bit_quant_type: fp4
|
| 2463 |
+
- bnb_4bit_use_double_quant: False
|
| 2464 |
+
- bnb_4bit_compute_dtype: float32
|
| 2465 |
+
|
| 2466 |
+
### Framework versions
|
| 2467 |
+
|
| 2468 |
+
|
| 2469 |
- 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 sha256:55bcb539a0e6b191941af31a3736816651b57886504b94a00762f5714e0a2c0d
|
| 3 |
size 21020682
|