7B Mistral Merges
Collection
A collection of my 7B parameter merges
•
3 items
•
Updated
🌠 Multiparadigm_7B
Multiparadigm_7B is a merge of the following models:
Quantizations
Thanks to mradermacher, static GGUF quants are available here.
Configuration
slices:
- sources:
- model: MTSAIR/multi_verse_model
layer_range: [0, 32]
- model: ResplendentAI/Paradigm_7B
layer_range: [0, 32]
merge_method: slerp
base_model: MTSAIR/multi_verse_model
parameters:
t:
- filter: self_attn
value: [0, 0.6, 0.3, 0.7, 1]
- filter: mlp
value: [1, 0.6, 0.7, 0.3, 0]
- value: 0.6
dtype: bfloat16
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 76.08 |
| AI2 Reasoning Challenge (25-Shot) | 73.21 |
| HellaSwag (10-Shot) | 88.95 |
| MMLU (5-Shot) | 64.28 |
| TruthfulQA (0-shot) | 76.87 |
| Winogrande (5-shot) | 83.82 |
| GSM8k (5-shot) | 69.37 |
- Downloads last month
- 12
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Model tree for rmdhirr/Multiparadigm_7B
Collection including rmdhirr/Multiparadigm_7B
Evaluation results
- normalized accuracy on AI2 Reasoning Challenge (25-Shot)test set Open LLM Leaderboard73.210
- normalized accuracy on HellaSwag (10-Shot)validation set Open LLM Leaderboard88.950
- accuracy on MMLU (5-Shot)test set Open LLM Leaderboard64.280
- mc2 on TruthfulQA (0-shot)validation set Open LLM Leaderboard76.870
- accuracy on Winogrande (5-shot)validation set Open LLM Leaderboard83.820
- accuracy on GSM8k (5-shot)test set Open LLM Leaderboard69.370