Isabelle Cudlitz
commited on
Commit
·
7fc3270
1
Parent(s):
9af8db5
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -1944,4 +1944,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 1944 |
### Framework versions
|
| 1945 |
|
| 1946 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1947 |
- PEFT 0.6.2
|
|
|
|
| 1944 |
### Framework versions
|
| 1945 |
|
| 1946 |
|
| 1947 |
+
- PEFT 0.6.2
|
| 1948 |
+
## Training procedure
|
| 1949 |
+
|
| 1950 |
+
|
| 1951 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 1952 |
+
- load_in_8bit: True
|
| 1953 |
+
- load_in_4bit: False
|
| 1954 |
+
- llm_int8_threshold: 6.0
|
| 1955 |
+
- llm_int8_skip_modules: None
|
| 1956 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 1957 |
+
- llm_int8_has_fp16_weight: False
|
| 1958 |
+
- bnb_4bit_quant_type: fp4
|
| 1959 |
+
- bnb_4bit_use_double_quant: False
|
| 1960 |
+
- bnb_4bit_compute_dtype: float32
|
| 1961 |
+
|
| 1962 |
+
### Framework versions
|
| 1963 |
+
|
| 1964 |
+
|
| 1965 |
- 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 sha256:14e3cde42f9d1d1be0b9b2547874fa4c6c936a4c0baf01b89d39cbc9c39425c3
|
| 3 |
size 21020682
|