Helios-10.7B-v2
This model is a merge of the following models made with mergekit:
🧩 Configuration
slices:
- sources:
- model: jeonsworld/CarbonVillain-en-10.7B-v2
layer_range: [0, 32]
- model: kekmodel/StopCarbon-10.7B-v5
layer_range: [0, 32]
merge_method: slerp
base_model: jeonsworld/CarbonVillain-en-10.7B-v4
parameters:
t:
- filter: self_attn
value: [0, 0.5, 0.3, 0.7, 1]
- filter: mlp
value: [1, 0.5, 0.7, 0.3, 0]
- value: 0.5
dtype: bfloat16
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
Metric | Value |
---|---|
Avg. | 42.25 |
AI2 Reasoning Challenge (25-Shot) | 39.16 |
HellaSwag (10-Shot) | 46.63 |
MMLU (5-Shot) | 41.57 |
TruthfulQA (0-shot) | 55.51 |
Winogrande (5-shot) | 70.64 |
GSM8k (5-shot) | 0.00 |
- Downloads last month
- 2,070
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
Model tree for occultml/Helios-10.7B-v2
Evaluation results
- normalized accuracy on AI2 Reasoning Challenge (25-Shot)test set Open LLM Leaderboard39.160
- normalized accuracy on HellaSwag (10-Shot)validation set Open LLM Leaderboard46.630
- accuracy on MMLU (5-Shot)test set Open LLM Leaderboard41.570
- mc2 on TruthfulQA (0-shot)validation set Open LLM Leaderboard55.510
- accuracy on Winogrande (5-shot)validation set Open LLM Leaderboard70.640
- accuracy on GSM8k (5-shot)test set Open LLM Leaderboard0.000