初始化项目,由ModelHub XC社区提供模型
Model: uukuguy/mistral-7b-platypus-fp16-dare-0.9 Source: Original Platform
This commit is contained in:
9
README.md
Normal file
9
README.md
Normal file
@@ -0,0 +1,9 @@
|
||||
---
|
||||
license: llama2
|
||||
---
|
||||
Experiment for DARE(Drop and REscale), most of the delta parameters can be directly set to zeros without affecting the capabilities of SFT LMs and larger models can tolerate a higher proportion of discarded parameters.
|
||||
|
||||
|
||||
| Model | Average | ARC | HellaSwag | MMLU | TruthfulQA | Winogrande | GSM8K | DROP |
|
||||
| ------ | ------ | ------ | ------ | ------ | ------ | ------ | ------ | ------ |
|
||||
| bhenrym14/mistral-7b-platypus-fp16 | 56.89 | 63.05 | 84.15 | 64.11 | 45.07 | 78.53 | 17.36 | 45.92 |
|
||||
Reference in New Issue
Block a user