Update README.md
Browse files
README.md
CHANGED
@@ -292,11 +292,12 @@ We initialized a fp16-precision ChatMusician-Base from the LLaMA2-7B-Base weight
|
|
292 |
was 2048. We utilized 16 80GB-A800 GPUs for one epoch pre-training and 8 32GB-V100 GPUs for two epoch fine-tuning. DeepSpeed was employed for memory efficiency, and the AdamW optimizer was used with a 1e-4 learning rate and a 5% warmup cosine scheduler. Gradient clipping was set at 1.0. The LoRA parameters dimension, alpha, and
|
293 |
dropout were set to 64, 16, and 0.1, with a batch size of 8.
|
294 |
|
|
|
295 |
## Evaluation
|
296 |
|
297 |
1. Music understanding abilities are evaluated on the [MusicTheoryBench](https://huggingface.co/datasets/m-a-p/MusicTheoryBench). The following figure is zero-shot accuracy on MusicTheoryBench.
|
298 |
We included GPT-3.5, GPT-4, LLaMA2-7B-Base, ChatMusician-Base, and ChatMusician. The blue bar represents the performance on the music knowledge metric, and the red bar represents the music reasoning metric. The dashed line corresponds to a random baseline, with a score of 25%.![MusicTheoryBench_result](./MusicTheoryBench_result_plt.png)
|
299 |
-
|
300 |
|
301 |
|
302 |
## Usage
|
|
|
292 |
was 2048. We utilized 16 80GB-A800 GPUs for one epoch pre-training and 8 32GB-V100 GPUs for two epoch fine-tuning. DeepSpeed was employed for memory efficiency, and the AdamW optimizer was used with a 1e-4 learning rate and a 5% warmup cosine scheduler. Gradient clipping was set at 1.0. The LoRA parameters dimension, alpha, and
|
293 |
dropout were set to 64, 16, and 0.1, with a batch size of 8.
|
294 |
|
295 |
+
|
296 |
## Evaluation
|
297 |
|
298 |
1. Music understanding abilities are evaluated on the [MusicTheoryBench](https://huggingface.co/datasets/m-a-p/MusicTheoryBench). The following figure is zero-shot accuracy on MusicTheoryBench.
|
299 |
We included GPT-3.5, GPT-4, LLaMA2-7B-Base, ChatMusician-Base, and ChatMusician. The blue bar represents the performance on the music knowledge metric, and the red bar represents the music reasoning metric. The dashed line corresponds to a random baseline, with a score of 25%.![MusicTheoryBench_result](./MusicTheoryBench_result_plt.png)
|
300 |
+
2. General language abilities of ChatMusician are evaluated on the [Massive Multitask Language Understanding (MMLU) dataset](https://huggingface.co/datasets/lukaemon/mmlu).
|
301 |
|
302 |
|
303 |
## Usage
|