KallistiTMR
commited on
Commit
·
4f40b4a
1
Parent(s):
cddf436
Upload model
Browse files- README.md +24 -0
- adapter_model.bin +1 -1
README.md
CHANGED
@@ -510,6 +510,28 @@ The following `bitsandbytes` quantization config was used during training:
|
|
510 |
- bnb_4bit_use_double_quant: False
|
511 |
- bnb_4bit_compute_dtype: float16
|
512 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
513 |
The following `bitsandbytes` quantization config was used during training:
|
514 |
- load_in_8bit: False
|
515 |
- load_in_4bit: True
|
@@ -568,5 +590,7 @@ The following `bitsandbytes` quantization config was used during training:
|
|
568 |
- PEFT 0.4.0
|
569 |
- PEFT 0.4.0
|
570 |
- PEFT 0.4.0
|
|
|
|
|
571 |
|
572 |
- PEFT 0.4.0
|
|
|
510 |
- bnb_4bit_use_double_quant: False
|
511 |
- bnb_4bit_compute_dtype: float16
|
512 |
|
513 |
+
The following `bitsandbytes` quantization config was used during training:
|
514 |
+
- load_in_8bit: False
|
515 |
+
- load_in_4bit: True
|
516 |
+
- llm_int8_threshold: 6.0
|
517 |
+
- llm_int8_skip_modules: None
|
518 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
519 |
+
- llm_int8_has_fp16_weight: False
|
520 |
+
- bnb_4bit_quant_type: nf4
|
521 |
+
- bnb_4bit_use_double_quant: False
|
522 |
+
- bnb_4bit_compute_dtype: float16
|
523 |
+
|
524 |
+
The following `bitsandbytes` quantization config was used during training:
|
525 |
+
- load_in_8bit: False
|
526 |
+
- load_in_4bit: True
|
527 |
+
- llm_int8_threshold: 6.0
|
528 |
+
- llm_int8_skip_modules: None
|
529 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
530 |
+
- llm_int8_has_fp16_weight: False
|
531 |
+
- bnb_4bit_quant_type: nf4
|
532 |
+
- bnb_4bit_use_double_quant: False
|
533 |
+
- bnb_4bit_compute_dtype: float16
|
534 |
+
|
535 |
The following `bitsandbytes` quantization config was used during training:
|
536 |
- load_in_8bit: False
|
537 |
- load_in_4bit: True
|
|
|
590 |
- PEFT 0.4.0
|
591 |
- PEFT 0.4.0
|
592 |
- PEFT 0.4.0
|
593 |
+
- PEFT 0.4.0
|
594 |
+
- PEFT 0.4.0
|
595 |
|
596 |
- 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 sha256:678bc429e56413483ed0ab4fb2fef7e2c00eabae823bc263d56fc9c3f76cc036
|
3 |
size 134263757
|