Hermes-low-tune-3.1 / README.md
nlpguy's picture
Adding Evaluation Results (#1)
f90bbae verified
---
tags:
- mergekit
- merge
base_model:
- teknium/OpenHermes-2.5-Mistral-7B
- nlpguy/Hermes-low-tune-2
- charlesdedampierre/TopicNeuralHermes-2.5-Mistral-7B
- openaccess-ai-collective/openhermes-2_5-dpo-no-robots
- flemmingmiguel/Mistrality-7B
- beowolx/MistralHermes-CodePro-7B-v1
license_name: apache-2.0-mit-dual-license
license_link: https://huggingface.co/nlpguy/Hermes-low-tune-3.1/raw/main/LICENSE
model-index:
- name: Hermes-low-tune-3.1
results:
- task:
type: text-generation
name: Text Generation
dataset:
name: AI2 Reasoning Challenge (25-Shot)
type: ai2_arc
config: ARC-Challenge
split: test
args:
num_few_shot: 25
metrics:
- type: acc_norm
value: 65.44
name: normalized accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nlpguy/Hermes-low-tune-3.1
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: HellaSwag (10-Shot)
type: hellaswag
split: validation
args:
num_few_shot: 10
metrics:
- type: acc_norm
value: 84.6
name: normalized accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nlpguy/Hermes-low-tune-3.1
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MMLU (5-Shot)
type: cais/mmlu
config: all
split: test
args:
num_few_shot: 5
metrics:
- type: acc
value: 64.13
name: accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nlpguy/Hermes-low-tune-3.1
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: TruthfulQA (0-shot)
type: truthful_qa
config: multiple_choice
split: validation
args:
num_few_shot: 0
metrics:
- type: mc2
value: 53.59
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nlpguy/Hermes-low-tune-3.1
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: Winogrande (5-shot)
type: winogrande
config: winogrande_xl
split: validation
args:
num_few_shot: 5
metrics:
- type: acc
value: 78.61
name: accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nlpguy/Hermes-low-tune-3.1
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: GSM8k (5-shot)
type: gsm8k
config: main
split: test
args:
num_few_shot: 5
metrics:
- type: acc
value: 63.46
name: accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nlpguy/Hermes-low-tune-3.1
name: Open LLM Leaderboard
---
# merged
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 [task arithmetic](https://arxiv.org/abs/2212.04089) merge method using [teknium/OpenHermes-2.5-Mistral-7B](https://huggingface.co/teknium/OpenHermes-2.5-Mistral-7B) as a base.
### Models Merged
The following models were included in the merge:
* [nlpguy/Hermes-low-tune-2](https://huggingface.co/nlpguy/Hermes-low-tune-2)
* [charlesdedampierre/TopicNeuralHermes-2.5-Mistral-7B](https://huggingface.co/charlesdedampierre/TopicNeuralHermes-2.5-Mistral-7B)
* [openaccess-ai-collective/openhermes-2_5-dpo-no-robots](https://huggingface.co/openaccess-ai-collective/openhermes-2_5-dpo-no-robots)
* [flemmingmiguel/Mistrality-7B](https://huggingface.co/flemmingmiguel/Mistrality-7B)
* [beowolx/MistralHermes-CodePro-7B-v1](https://huggingface.co/beowolx/MistralHermes-CodePro-7B-v1)
### Configuration
The following YAML configuration was used to produce this model:
```yaml
base_model: teknium/OpenHermes-2.5-Mistral-7B
dtype: bfloat16
merge_method: task_arithmetic
slices:
- sources:
- layer_range: [0, 32]
model: teknium/OpenHermes-2.5-Mistral-7B
- layer_range: [0, 32]
model: nlpguy/Hermes-low-tune-2
parameters:
weight: 0.2
- layer_range: [0, 32]
model: beowolx/MistralHermes-CodePro-7B-v1
parameters:
weight: 0.2
- layer_range: [0, 32]
model: flemmingmiguel/Mistrality-7B
parameters:
weight: 0.2
- layer_range: [0, 32]
model: charlesdedampierre/TopicNeuralHermes-2.5-Mistral-7B
parameters:
weight: 0.2
- layer_range: [0, 32]
model: openaccess-ai-collective/openhermes-2_5-dpo-no-robots
parameters:
weight: 0.2
```
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_nlpguy__Hermes-low-tune-3.1)
| Metric |Value|
|---------------------------------|----:|
|Avg. |68.31|
|AI2 Reasoning Challenge (25-Shot)|65.44|
|HellaSwag (10-Shot) |84.60|
|MMLU (5-Shot) |64.13|
|TruthfulQA (0-shot) |53.59|
|Winogrande (5-shot) |78.61|
|GSM8k (5-shot) |63.46|