Original model description:

license: cc-by-nc-4.0

Replaced Zephyr by Airoboros 2.2 and OpenOrca by SynthIA in the mix, the reason why is to see if using merged Mistral models using all the same prompt format would be a better step or not.

Description

This repo contains fp16 files of Mistral-11B-SynthIAirOmniMix.

Model used

Prompt template

3 out of 4 models use the same prompting format in this merge.

The best one should be this one, since Zephyr and OpenOrca is out of the merge:

(SYSTEM: {context}) - Not mandatory
USER: {prompt}
ASSISTANT:

But this one (maybe) work too:

Below is an instruction that describes a task. Write a response that appropriately completes the request.

### Instruction:
{prompt}

### Response:

The secret sauce

Mistral-11B-SynthIAOpenPlatypus :

slices:
  - sources:
    - model: "/content/drive/MyDrive/SynthIA-7B-v1.5-bf16"
      layer_range: [0, 24]
  - sources:
    - model: akjindal53244/Mistral-7B-v0.1-Open-Platypus
      layer_range: [8, 32]
merge_method: passthrough
dtype: bfloat16

Mistral-11B-CC-Airo :

slices:
  - sources:
    - model: "/content/drive/MyDrive/CC-v1.1-7B-bf16"
      layer_range: [0, 24]
  - sources:
    - model: "/content/drive/MyDrive/Mistral-7B-Airoboros-2.2-bf16"
      layer_range: [8, 32]
merge_method: passthrough
dtype: bfloat16

Mistral-11B-SynthIAirOmniMix :

slices:
  - sources:
      - model: Mistral-11B-SynthIAOpenPlatypus
        layer_range: [0, 48]
      - model: Mistral-11B-CC-Airo
        layer_range: [0, 48]
merge_method: slerp
base_model: Mistral-11B-OpenOrcaPlatypus
parameters:
  t:
    - filter: lm_head 
      value: [0.75]
    - filter: embed_tokens
      value: [0.75]
    - filter: self_attn
      value: [0.75, 0.25]
    - filter: mlp
      value:  [0.25, 0.75]
    - filter: layernorm
      value: [0.5, 0.5]
    - filter: modelnorm
      value: [0.75]
    - value: 0.5 # fallback for rest of tensors
dtype: bfloat16

I use mergekit for all the manipulation told here.

Some scoring I done myself

Task	Version	Metric	Value		Stderr
arc_challenge	0	acc	0.5410	±	0.0146
		acc_norm	0.5640	±	0.0145
arc_easy	0	acc	0.8228	±	0.0078
		acc_norm	0.8068	±	0.0081
hellaswag	0	acc	0.6274	±	0.0048
		acc_norm	0.8167	±	0.0039
piqa	0	acc	0.8052	±	0.0092
		acc_norm	0.8232	±	0.0089
truthfulqa_mc	1	mc1	0.3905	±	0.0171
		mc2	0.5592	±	0.0155
winogrande	0	acc	0.7364	±	0.0124

Others

Special thanks to Sushi, Henky for the machine he give me for big task, and Charles Goddard for his amazing tool.

If you want to support me, you can here.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	54.56
ARC (25-shot)	62.46
HellaSwag (10-shot)	83.13
MMLU (5-shot)	63.47
TruthfulQA (0-shot)	55.69
Winogrande (5-shot)	76.4
GSM8K (5-shot)	11.9
DROP (3-shot)	28.88