Commit
•
327aac8
1
Parent(s):
5ab3f47
Update README.md
Browse files
README.md
CHANGED
|
@@ -6,7 +6,7 @@ language:
|
|
| 6 |
|
| 7 |
# Hyacinth6B: A Trandidional Chinese Large Language Model
|
| 8 |
|
| 9 |
-
|
| 10 |
|
| 11 |
Hyacinth6B is a Tranditional Chinese Large Language Model which fine-tune from [chatglm3-base](https://huggingface.co/THUDM/chatglm3-6b-base),our goal is to find a balance between model lightness and performance, striving to maximize performance while using a comparatively lightweight model. Hyacinth6B was developed with this objective in mind, aiming to fully leverage the core capabilities of LLMs without incurring substantial resource costs, effectively pushing the boundaries of smaller models' performance. The training approach involves parameter-efficient fine-tuning using the Low-Rank Adaptation (LoRA) method.
|
| 12 |
At last, we evaluated Hyacinth6B, examining its performance across various aspects. Hyacinth6B shows commendable performance in certain metrics, even surpassing ChatGPT in two categories. We look forward to providing more resources and possibilities for the field of Traditional Chinese language processing. This research aims to expand the research scope of Traditional Chinese language models and enhance their applicability in different scenarios.
|
|
@@ -31,15 +31,15 @@ Training required approximately 20.6GB of VRAM without any quantization (default
|
|
| 31 |
|
| 32 |
# Evaluate Results
|
| 33 |
## CMMLU
|
| 34 |
-
|
| 35 |
## C-eval
|
| 36 |
-
|
| 37 |
## TC-eval by MediaTek Research
|
| 38 |
-
|
| 39 |
## MT-bench
|
| 40 |
-
|
| 41 |
## LLM-eval by NTU Miu Lab
|
| 42 |
-
|
| 43 |
## Bailong Bench
|
| 44 |
|
| 45 |
|
|
|
|
| 6 |
|
| 7 |
# Hyacinth6B: A Trandidional Chinese Large Language Model
|
| 8 |
|
| 9 |
+
<img src="./pics/hyacinth.jpeg" alt="image_name png"/>
|
| 10 |
|
| 11 |
Hyacinth6B is a Tranditional Chinese Large Language Model which fine-tune from [chatglm3-base](https://huggingface.co/THUDM/chatglm3-6b-base),our goal is to find a balance between model lightness and performance, striving to maximize performance while using a comparatively lightweight model. Hyacinth6B was developed with this objective in mind, aiming to fully leverage the core capabilities of LLMs without incurring substantial resource costs, effectively pushing the boundaries of smaller models' performance. The training approach involves parameter-efficient fine-tuning using the Low-Rank Adaptation (LoRA) method.
|
| 12 |
At last, we evaluated Hyacinth6B, examining its performance across various aspects. Hyacinth6B shows commendable performance in certain metrics, even surpassing ChatGPT in two categories. We look forward to providing more resources and possibilities for the field of Traditional Chinese language processing. This research aims to expand the research scope of Traditional Chinese language models and enhance their applicability in different scenarios.
|
|
|
|
| 31 |
|
| 32 |
# Evaluate Results
|
| 33 |
## CMMLU
|
| 34 |
+
<img src="./pics/cmmlu.png" alt="image_name png"/>
|
| 35 |
## C-eval
|
| 36 |
+
<img src="./pics/ceval.png" alt="image_name png"/>
|
| 37 |
## TC-eval by MediaTek Research
|
| 38 |
+
<img src="./pics/tc-eval.png" alt="image_name png"/>
|
| 39 |
## MT-bench
|
| 40 |
+
<img src="./pics/dashB.png" alt="image_name png"/>
|
| 41 |
## LLM-eval by NTU Miu Lab
|
| 42 |
+
<img src="./pics/llmeval.png" alt="image_name png"/>
|
| 43 |
## Bailong Bench
|
| 44 |
|
| 45 |
|