Best model with eval_loss: 0.6388353109359741
Browse files- README.md +82 -0
- adapter_model.safetensors +1 -1
README.md
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: Bllossom/llama-3.2-Korean-Bllossom-AICA-5B
|
| 3 |
+
library_name: peft
|
| 4 |
+
license: llama3.2
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
model-index:
|
| 8 |
+
- name: HGU_rulebook-Llama3.2-Bllossom-5B_fine-tuning-QLoRA-32_64_3
|
| 9 |
+
results: []
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 13 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 14 |
+
|
| 15 |
+
# HGU_rulebook-Llama3.2-Bllossom-5B_fine-tuning-QLoRA-32_64_3
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [Bllossom/llama-3.2-Korean-Bllossom-AICA-5B](https://huggingface.co/Bllossom/llama-3.2-Korean-Bllossom-AICA-5B) on an unknown dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 5.6952
|
| 20 |
+
|
| 21 |
+
## Model description
|
| 22 |
+
|
| 23 |
+
More information needed
|
| 24 |
+
|
| 25 |
+
## Intended uses & limitations
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Training and evaluation data
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training procedure
|
| 34 |
+
|
| 35 |
+
### Training hyperparameters
|
| 36 |
+
|
| 37 |
+
The following hyperparameters were used during training:
|
| 38 |
+
- learning_rate: 1e-05
|
| 39 |
+
- train_batch_size: 2
|
| 40 |
+
- eval_batch_size: 2
|
| 41 |
+
- seed: 42
|
| 42 |
+
- gradient_accumulation_steps: 8
|
| 43 |
+
- total_train_batch_size: 16
|
| 44 |
+
- optimizer: Use adamw_bnb_8bit with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 45 |
+
- lr_scheduler_type: cosine_with_restarts
|
| 46 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 47 |
+
- training_steps: 942
|
| 48 |
+
- mixed_precision_training: Native AMP
|
| 49 |
+
|
| 50 |
+
### Training results
|
| 51 |
+
|
| 52 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 53 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 54 |
+
| 12.459 | 0.2991 | 47 | 11.8428 |
|
| 55 |
+
| 8.9527 | 0.5982 | 94 | 8.3608 |
|
| 56 |
+
| 6.8438 | 0.8974 | 141 | 6.6652 |
|
| 57 |
+
| 6.1272 | 1.1965 | 188 | 6.0176 |
|
| 58 |
+
| 5.858 | 1.4956 | 235 | 5.8316 |
|
| 59 |
+
| 5.7707 | 1.7947 | 282 | 5.7618 |
|
| 60 |
+
| 5.7435 | 2.0939 | 329 | 5.7331 |
|
| 61 |
+
| 5.7215 | 2.3930 | 376 | 5.7194 |
|
| 62 |
+
| 5.7118 | 2.6921 | 423 | 5.7114 |
|
| 63 |
+
| 5.7047 | 2.9912 | 470 | 5.7063 |
|
| 64 |
+
| 5.703 | 3.2904 | 517 | 5.7027 |
|
| 65 |
+
| 5.6973 | 3.5895 | 564 | 5.7001 |
|
| 66 |
+
| 5.6959 | 3.8886 | 611 | 5.6985 |
|
| 67 |
+
| 5.6921 | 4.1877 | 658 | 5.6972 |
|
| 68 |
+
| 5.6968 | 4.4869 | 705 | 5.6965 |
|
| 69 |
+
| 5.6954 | 4.7860 | 752 | 5.6959 |
|
| 70 |
+
| 5.6958 | 5.0851 | 799 | 5.6955 |
|
| 71 |
+
| 5.6937 | 5.3842 | 846 | 5.6953 |
|
| 72 |
+
| 5.6915 | 5.6834 | 893 | 5.6952 |
|
| 73 |
+
| 5.6951 | 5.9825 | 940 | 5.6952 |
|
| 74 |
+
|
| 75 |
+
|
| 76 |
+
### Framework versions
|
| 77 |
+
|
| 78 |
+
- PEFT 0.12.0
|
| 79 |
+
- Transformers 4.46.2
|
| 80 |
+
- Pytorch 2.0.1+cu118
|
| 81 |
+
- Datasets 3.0.0
|
| 82 |
+
- Tokenizers 0.20.1
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 72135352
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8f7399cc9e6fc3ef753412680e3f47aa1d344c8b1f68503cfcb21db67a91a759
|
| 3 |
size 72135352
|