KallistiTMR
commited on
Commit
•
4e3062e
1
Parent(s):
443aec2
Upload model
Browse files- README.md +24 -0
- adapter_model.bin +1 -1
README.md
CHANGED
@@ -466,6 +466,28 @@ The following `bitsandbytes` quantization config was used during training:
|
|
466 |
- bnb_4bit_use_double_quant: False
|
467 |
- bnb_4bit_compute_dtype: float16
|
468 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
469 |
The following `bitsandbytes` quantization config was used during training:
|
470 |
- load_in_8bit: False
|
471 |
- load_in_4bit: True
|
@@ -520,5 +542,7 @@ The following `bitsandbytes` quantization config was used during training:
|
|
520 |
- PEFT 0.4.0
|
521 |
- PEFT 0.4.0
|
522 |
- PEFT 0.4.0
|
|
|
|
|
523 |
|
524 |
- PEFT 0.4.0
|
|
|
466 |
- bnb_4bit_use_double_quant: False
|
467 |
- bnb_4bit_compute_dtype: float16
|
468 |
|
469 |
+
The following `bitsandbytes` quantization config was used during training:
|
470 |
+
- load_in_8bit: False
|
471 |
+
- load_in_4bit: True
|
472 |
+
- llm_int8_threshold: 6.0
|
473 |
+
- llm_int8_skip_modules: None
|
474 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
475 |
+
- llm_int8_has_fp16_weight: False
|
476 |
+
- bnb_4bit_quant_type: nf4
|
477 |
+
- bnb_4bit_use_double_quant: False
|
478 |
+
- bnb_4bit_compute_dtype: float16
|
479 |
+
|
480 |
+
The following `bitsandbytes` quantization config was used during training:
|
481 |
+
- load_in_8bit: False
|
482 |
+
- load_in_4bit: True
|
483 |
+
- llm_int8_threshold: 6.0
|
484 |
+
- llm_int8_skip_modules: None
|
485 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
486 |
+
- llm_int8_has_fp16_weight: False
|
487 |
+
- bnb_4bit_quant_type: nf4
|
488 |
+
- bnb_4bit_use_double_quant: False
|
489 |
+
- bnb_4bit_compute_dtype: float16
|
490 |
+
|
491 |
The following `bitsandbytes` quantization config was used during training:
|
492 |
- load_in_8bit: False
|
493 |
- load_in_4bit: True
|
|
|
542 |
- PEFT 0.4.0
|
543 |
- PEFT 0.4.0
|
544 |
- PEFT 0.4.0
|
545 |
+
- PEFT 0.4.0
|
546 |
+
- PEFT 0.4.0
|
547 |
|
548 |
- PEFT 0.4.0
|
adapter_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 134263757
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:276ba1d60a2149a72e3679bcef549f2687e688f7dd27e79a524b41d414d90998
|
3 |
size 134263757
|