Merges
Collection
Personal Merges
•
97 items
•
Updated
•
1
merge
This is a merge of pre-trained language models created using mergekit.
Model details:
V0.1 was spitting out nonsense, so have attempted a different merge method and parameters.
Merge Details
Merge Method
This model was merged using the TIES merge method using Gryphe/Pantheon-RP-1.6.1-12b-Nemo as a base.
Models Merged
The following models were included in the merge:
Configuration
The following YAML configuration was used to produce this model:
models:
- model: spow12/ChatWaifu_12B_v2.0
parameters:
density: 0.25
weight: 0.25
- model: Gryphe/Pantheon-RP-1.6.1-12b-Nemo
parameters:
density: 0.5
weight: 0.5
merge_method: ties
base_model: Gryphe/Pantheon-RP-1.6.1-12b-Nemo
parameters:
normalize: false
int8_mask: true
dtype: float16
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 20.81 |
| IFEval (0-Shot) | 26.83 |
| BBH (3-Shot) | 36.74 |
| MATH Lvl 5 (4-Shot) | 5.44 |
| GPQA (0-shot) | 9.06 |
| MuSR (0-shot) | 19.64 |
| MMLU-PRO (5-shot) | 27.14 |
- Downloads last month
- 25
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
Model tree for Triangle104/Pantheon_ChatWaifu_V0.2
Merge model
this model
Collection including Triangle104/Pantheon_ChatWaifu_V0.2
Evaluation results
- strict accuracy on IFEval (0-Shot)Open LLM Leaderboard26.830
- normalized accuracy on BBH (3-Shot)Open LLM Leaderboard36.740
- exact match on MATH Lvl 5 (4-Shot)Open LLM Leaderboard5.440
- acc_norm on GPQA (0-shot)Open LLM Leaderboard9.060
- acc_norm on MuSR (0-shot)Open LLM Leaderboard19.640
- accuracy on MMLU-PRO (5-shot)test set Open LLM Leaderboard27.140