Commit ·
9dd0c7a
1
Parent(s): ac66b14
Model save
Browse files- README.md +27 -2
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -5,6 +5,7 @@ tags:
|
|
| 5 |
model-index:
|
| 6 |
- name: lora-flan-t5-small
|
| 7 |
results: []
|
|
|
|
| 8 |
---
|
| 9 |
|
| 10 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -28,6 +29,28 @@ More information needed
|
|
| 28 |
|
| 29 |
## Training procedure
|
| 30 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 31 |
### Training hyperparameters
|
| 32 |
|
| 33 |
The following hyperparameters were used during training:
|
|
@@ -37,8 +60,8 @@ The following hyperparameters were used during training:
|
|
| 37 |
- seed: 42
|
| 38 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 39 |
- lr_scheduler_type: linear
|
| 40 |
-
- lr_scheduler_warmup_ratio: 0.
|
| 41 |
-
- training_steps:
|
| 42 |
|
| 43 |
### Training results
|
| 44 |
|
|
@@ -46,6 +69,8 @@ The following hyperparameters were used during training:
|
|
| 46 |
|
| 47 |
### Framework versions
|
| 48 |
|
|
|
|
|
|
|
| 49 |
- Transformers 4.30.0
|
| 50 |
- Pytorch 2.0.1+cu118
|
| 51 |
- Datasets 2.14.5
|
|
|
|
| 5 |
model-index:
|
| 6 |
- name: lora-flan-t5-small
|
| 7 |
results: []
|
| 8 |
+
library_name: peft
|
| 9 |
---
|
| 10 |
|
| 11 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 29 |
|
| 30 |
## Training procedure
|
| 31 |
|
| 32 |
+
|
| 33 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 34 |
+
- load_in_8bit: True
|
| 35 |
+
- load_in_4bit: False
|
| 36 |
+
- llm_int8_threshold: 6.0
|
| 37 |
+
- llm_int8_skip_modules: None
|
| 38 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 39 |
+
- llm_int8_has_fp16_weight: False
|
| 40 |
+
- bnb_4bit_quant_type: fp4
|
| 41 |
+
- bnb_4bit_use_double_quant: False
|
| 42 |
+
- bnb_4bit_compute_dtype: float32
|
| 43 |
+
|
| 44 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 45 |
+
- load_in_8bit: True
|
| 46 |
+
- load_in_4bit: False
|
| 47 |
+
- llm_int8_threshold: 6.0
|
| 48 |
+
- llm_int8_skip_modules: None
|
| 49 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 50 |
+
- llm_int8_has_fp16_weight: False
|
| 51 |
+
- bnb_4bit_quant_type: fp4
|
| 52 |
+
- bnb_4bit_use_double_quant: False
|
| 53 |
+
- bnb_4bit_compute_dtype: float32
|
| 54 |
### Training hyperparameters
|
| 55 |
|
| 56 |
The following hyperparameters were used during training:
|
|
|
|
| 60 |
- seed: 42
|
| 61 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 62 |
- lr_scheduler_type: linear
|
| 63 |
+
- lr_scheduler_warmup_ratio: 0.04
|
| 64 |
+
- training_steps: 15000
|
| 65 |
|
| 66 |
### Training results
|
| 67 |
|
|
|
|
| 69 |
|
| 70 |
### Framework versions
|
| 71 |
|
| 72 |
+
- PEFT 0.6.0.dev0
|
| 73 |
+
- PEFT 0.6.0.dev0
|
| 74 |
- Transformers 4.30.0
|
| 75 |
- Pytorch 2.0.1+cu118
|
| 76 |
- Datasets 2.14.5
|
adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2787473
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:996b46471c4331c519e5fddb72aed8231779463fd0b1d3cb5db1182cf8b367c9
|
| 3 |
size 2787473
|