File size: 4,054 Bytes
d11defc 7195ee3 d11defc 1cd5dde d11defc 1cd5dde d11defc fd52fe0 d11defc c1789f8 d11defc |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 |
---
license: cc-by-nc-4.0
---
This model should be fixed, it was MEANT to be BF16.
Don't mind this one at the moment, I need to finetune it for RP, it's just a test.
## Description
This repo contains fp16 files of Mistral-11B-OmniMix-bf16.
My goal for this model was only to make it score the highest possible with merge and layer toying, proving that:
- Benchmark are objective
- You should try a model yourself and don't go blindly to the highest rated one
- Merge/Layer toying CAN be usable to do better model (maybe?)
## Model used
- [Mistral-7B-OpenOrca](https://huggingface.co/Open-Orca/Mistral-7B-OpenOrca)
- [Mistral-7B-v0.1-Open-Platypus](https://huggingface.co/akjindal53244/Mistral-7B-v0.1-Open-Platypus)
- [CollectiveCognition-v1.1-Mistral-7B](https://huggingface.co/teknium/CollectiveCognition-v1.1-Mistral-7B)
- [zephyr-7b-alpha](https://huggingface.co/HuggingFaceH4/zephyr-7b-alpha)
## Prompt template
The best one after further testing is this one:
```
<|system|>
Below is an instruction that describes a task. Write a response that appropriately completes the request.
<|user|>
{prompt}
<|assistant|>
```
But these one work too:
```
Below is an instruction that describes a task. Write a response that appropriately completes the request.
### Instruction:
{prompt}
### Response:
```
```
USER: <prompt>
ASSISTANT:
```
Or use any prompting system from one of the 4 source model, should work.
## The secret sauce
Mistral-11B-OpenOrcaPlatypus :
```
slices:
- sources:
- model: Open-Orca/Mistral-7B-OpenOrca
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-Zephyr :
```
slices:
- sources:
- model: "/content/drive/MyDrive/CC-v1.1-7B-bf16"
layer_range: [0, 24]
- sources:
- model: "/content/drive/MyDrive/Zephyr-7B"
layer_range: [8, 32]
merge_method: passthrough
dtype: bfloat16
```
Mistral-11B-OmniMix :
```
slices:
- sources:
- model: Mistral-11B-OpenOrcaPlatypus
layer_range: [0, 48]
- model: Mistral-11B-CC-Zephyr
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](https://github.com/cg123/mergekit) for all the manipulation told here.
## Some scoring I done myself
hf-causal-experimental (pretrained=/content/drive/MyDrive/Mistral-11B-OmniMix-bf16), limit: None, provide_description: False, num_fewshot: 0, batch_size: 4
| Task |Version| Metric |Value | |Stderr|
|-------------|------:|--------|-----:|---|-----:|
|arc_challenge| 0|acc |0.5580|± |0.0145|
| | |acc_norm|0.5819|± |0.0144|
|arc_easy | 0|acc |0.8300|± |0.0077|
| | |acc_norm|0.8211|± |0.0079|
|hellaswag | 0|acc |0.6372|± |0.0048|
| | |acc_norm|0.8209|± |0.0038|
|piqa | 0|acc |0.8145|± |0.0091|
| | |acc_norm|0.8286|± |0.0088|
|truthfulqa_mc| 1|mc1 |0.3978|± |0.0171|
| | |mc2 |0.5680|± |0.0155|
|winogrande | 0|acc |0.7427|± |0.0123|
## Others
Special thanks to Sushi, [Henky](https://github.com/KoboldAI/KoboldAI-Client) for the machine he give me for big task, and [Charles Goddard](https://github.com/cg123) for his amazing tool.
If you want to support me, you can [here](https://ko-fi.com/undiai).
|