mpasila's picture
Clarified quantization type
ef3c1ca verified
---
license: cc-by-nc-4.0
tags:
- mergekit
- merge
base_model:
- BlueNipples/SnowLotus-v2-10.7B
- Himitsui/KuroMitsu-11B
model-index:
- name: Kuro-Lotus-10.7B
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: 68.69
name: normalized accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=saishf/Kuro-Lotus-10.7B
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: 87.51
name: normalized accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=saishf/Kuro-Lotus-10.7B
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: 66.64
name: accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=saishf/Kuro-Lotus-10.7B
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: 58.27
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=saishf/Kuro-Lotus-10.7B
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: 84.21
name: accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=saishf/Kuro-Lotus-10.7B
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: 66.11
name: accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=saishf/Kuro-Lotus-10.7B
name: Open LLM Leaderboard
---
This is an ExLlamaV2 quantized model of [saishf/Kuro-Lotus-10.7B](https://huggingface.co/saishf/Kuro-Lotus-10.7B) using the default calibration dataset.
The quants are uploaded on individual branches and the list is here:
[4bpw](https://huggingface.co/mpasila/Kuro-Lotus-10.7B-exl2/tree/4bpw)
[3.75bpw](https://huggingface.co/mpasila/Kuro-Lotus-10.7B-exl2/tree/3.75bpw)
[3.5bpw](https://huggingface.co/mpasila/Kuro-Lotus-10.7B-exl2/tree/3.5bpw)
[3.25bpw](https://huggingface.co/mpasila/Kuro-Lotus-10.7B-exl2/tree/3.25bpw)
[3bpw](https://huggingface.co/mpasila/Kuro-Lotus-10.7B-exl2/tree/3bpw)
Prompt format is Alpaca.
# Original Model card
# merge
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 SLERP merge method.
### Models Merged
The following models were included in the merge:
* [BlueNipples/SnowLotus-v2-10.7B](https://huggingface.co/BlueNipples/SnowLotus-v2-10.7B)
* [Himitsui/KuroMitsu-11B](https://huggingface.co/Himitsui/KuroMitsu-11B)
### Configuration
The following YAML configuration was used to produce this model:
```yaml
slices:
- sources:
- model: Himitsui/KuroMitsu-11B
layer_range: [0, 48]
- model: BlueNipples/SnowLotus-v2-10.7B
layer_range: [0, 48]
merge_method: slerp
base_model: Himitsui/KuroMitsu-11B
parameters:
t:
- filter: self_attn
value: [0.6, 0.7, 0.8, 0.9, 1]
- filter: mlp
value: [0.4, 0.3, 0.2, 0.1, 0]
- value: 0.5
dtype: bfloat16
```
# [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_saishf__Kuro-Lotus-10.7B)
| Metric |Value|
|---------------------------------|----:|
|Avg. |71.90|
|AI2 Reasoning Challenge (25-Shot)|68.69|
|HellaSwag (10-Shot) |87.51|
|MMLU (5-Shot) |66.64|
|TruthfulQA (0-shot) |58.27|
|Winogrande (5-shot) |84.21|
|GSM8k (5-shot) |66.11|