File size: 7,463 Bytes
cf816a1 a79253e cf816a1 ed346db cf816a1 ed346db cf816a1 ed4af49 cf816a1 ed4af49 cf816a1 a79253e 9e1cf47 a79253e 9e1cf47 a79253e 8bd317a a79253e |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 |
---
base_model:
- mistralai/Mistral-7B-v0.1
- berkeley-nest/Starling-LM-7B-alpha
- mlabonne/AlphaMonarch-7B
- cognitivecomputations/WestLake-7B-v2-laser
- senseable/garten2-7b
library_name: transformers
tags:
- mergekit
- merge
license: cc-by-nc-4.0
---
# Starling_Monarch_Westlake_Garten-7B-v0.1
After experimenting with density for a previous merge (containing similar models), I decided to experiment with weight gradients. My thought that was that if the merge was done with care and attention, I'd be able to create something greater than the sum of its parts.
Hoping that, through a merge of really good models, I'd be able to create soemthing greater than the sum of its parts.
I came across the EQ-Bench Benchmark [(Paper)](https://arxiv.org/abs/2312.06281) as part of my earlier testing. It is a very light and quick benchmark that yields powerful insights into how well the model performs in emotional intelligence related prompts.
As part of this process, I tried to figure out if there was a way to determine an optimal set of gradient weights that would lead to the most successful merge as measured against EQ-Bench. At first, my goal was to simply exceed WestLake-7B, but then I kept pushing to see what I could come up with.
Way too late in the process, did I learn that [dare_ties](https://arxiv.org/abs/2311.03099) has a random element to it, but considered it valuable information for next time. After concluding that project, I began collecting more data, this time setting a specified seed in mergekit for reproducibility.
This model is *not* a result of the above work but is the genesis of how this model came to be.
I present, Starling_Monarch_Westlake_Garten-7B-v0.1, the only 7B model to score > 80 on the EQ-Bench v2.1 benchmark found [here](https://github.com/EQ-bench/EQ-Bench), outscoring larger models like [abacusai/Smaug-72B-v0.1](https://huggingface.co/abacusai/Smaug-72B-v0.1) and [cognitivecomputations/dolphin-2.2-70b](https://huggingface.co/cognitivecomputations/dolphin-2.2-70b)
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
## Merge Details
### Merge Method
This model was merged using the [DARE](https://arxiv.org/abs/2311.03099) [TIES](https://arxiv.org/abs/2306.01708) merge method using [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) as a base.
The seed for this merge is 176
### Models Merged
The following models were included in the merge:
* [berkeley-nest/Starling-LM-7B-alpha](https://huggingface.co/berkeley-nest/Starling-LM-7B-alpha)
* [mlabonne/AlphaMonarch-7B](https://huggingface.co/mlabonne/AlphaMonarch-7B)
* [cognitivecomputations/WestLake-7B-v2-laser](https://huggingface.co/cognitivecomputations/WestLake-7B-v2-laser)
* [senseable/garten2-7b](https://huggingface.co/senseable/garten2-7b)
### Configuration
The following YAML configuration was used to produce this model:
```yaml
models:
- model: mistralai/Mistral-7B-v0.1
# No parameters necessary for base model
- model: cognitivecomputations/WestLake-7B-v2-laser
parameters:
density: 0.58
weight: [0.3877, 0.1636, 0.186, 0.0502]
- model: senseable/garten2-7b
parameters:
density: 0.58
weight: [0.234, 0.2423, 0.2148, 0.2775]
- model: berkeley-nest/Starling-LM-7B-alpha
parameters:
density: 0.58
weight: [0.1593, 0.1573, 0.1693, 0.3413]
- model: mlabonne/AlphaMonarch-7B
parameters:
density: 0.58
weight: [0.219, 0.4368, 0.4299, 0.331]
merge_method: dare_ties
base_model: mistralai/Mistral-7B-v0.1
parameters:
int8_mask: true
dtype: bfloat16
```
### Table of Benchmarks
## Open LLM Leaderboard
| | Average | ARC | HellaSwag | MMLU | TruthfulQA | Winogrande | GSM8K |
|---------------------------------------------------------|---------|-------|-----------|-------|------------|------------|-------|
| giraffe176/Starling_Monarch_Westlake_Garten-7B-v0.1 | XX.XX | XX.XX | XX.XX | XX.XX | XX.XX | XX.XX | XX.XX |
| mlabonne/AlphaMonarch-7B | 75.99 | 73.04 | 89.18 | 64.4 | 77.91 | 84.69 | 66.72 |
| senseable/WestLake-7B-v2 | 74.68 | 73.04 | 88.65 | 64.71 | 67.06 | 86.98 | 67.63 |
| berkeley-nest/Starling-LM-7B-alpha | 67.13 | 63.82 | 84.9 | 63.64 | 46.39 | 80.58 | 62.4 |
| senseable/garten2-7b | 72.65 | 69.37 | 87.54 | 65.44 | 59.5 | 84.69 | 69.37 |
## Yet Another LLM Leaderboard benchmarks
| Model |AGIEval|GPT4All|TruthfulQA|Bigbench|Average|
|---------------------------------------------------------------------------------------------------------------------------------|------:|------:|---------:|-------:|------:|
|[giraffe176/Starling_Monarch_Westlake_Garten-7B-v0.1](https://huggingface.co/giraffe176/Starling_Monarch_Westlake_Garten-7B-v0.1)| XX.XX| XX.XX| XX.XX| XX.XX| XX.XX|
|[mlabonne/AlphaMonarch-7B](https://huggingface.co/mlabonne/AlphaMonarch-7B) | 62.74| 45.37| 77.01| 78.39| 50.2 |
|[berkeley-nest/Starling-LM-7B-alpha](https://huggingface.co/senseable/garten2-7b) | 51.16| 42.06| 72.72| 47.33| 42.53|
## Misc. Benchmarks
| | MT-Bench | EQ-Bench v2.1 |
|---------------------------------------------------------|---------------------------------------------|---------------------------------------------------------------------------------|
| giraffe176/Starling_Monarch_Westlake_Garten-7B-v0.1 | 8.109375 | 80.01 (3 Shot, ooba) |
| mlabonne/AlphaMonarch-7B | 7.928125 | 76.08 |
| senseable/WestLake-7B-v2 | X | 78.7 |
| berkeley-nest/Starling-LM-7B-alpha | 8.09 | XX.X |
| senseable/garten2-7b | X | XX.X |
| claude-v1 | 7.900000 | 76.83 |
| gpt-3.5-turbo | 7.943750 | 71.74 |
| | [(Paper)](https://arxiv.org/abs/2306.05685) | [(Paper)](https://arxiv.org/abs/2312.06281) [Leaderboard](https://eqbench.com/) |
|