File size: 2,366 Bytes
093390c
 
20f6b45
 
 
 
 
 
 
 
 
093390c
20f6b45
 
 
 
 
 
 
 
 
 
 
 
 
af3b6da
 
47a5664
20f6b45
47a5664
20f6b45
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
32c49d3
 
 
 
 
 
 
 
 
 
 
20f6b45
32c49d3
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
---
license: mit
tags:
- merge
- mergekit
- lazymergekit
- rhysjones/phi-2-orange
- cognitivecomputations/dolphin-2_6-phi-2
base_model:
- rhysjones/phi-2-orange
- cognitivecomputations/dolphin-2_6-phi-2
---

# Phi-2-psy

Phi-2-psy is a merge of the following models:
* [rhysjones/phi-2-orange](https://huggingface.co/rhysjones/phi-2-orange)
* [cognitivecomputations/dolphin-2_6-phi-2](https://huggingface.co/cognitivecomputations/dolphin-2_6-phi-2)

## 🏆 Evaluation

The evaluation was performed using [LLM AutoEval](https://github.com/mlabonne/llm-autoeval) on Nous suite.

|                             Model                              |AGIEval|GPT4All|TruthfulQA|Bigbench|Average|
|----------------------------------------------------------------|------:|------:|---------:|-------:|------:|
|[**phi-2-psy**](https://huggingface.co/vince62s/phi-2-psy)|   **34.4**|  **71.4**|     **48.2**|   **38.1**|  **48.02**|
|[phixtral-2x2_8](https://huggingface.co/mlabonne/phixtral-2x2_8)|   34.1|  70.4|     48.8|   37.8|  47.78|
|[dolphin-2_6-phi-2](https://huggingface.co/cognitivecomputations/dolphin-2_6-phi-2)|  33.1|  69.9|     47.4|    37.2|  46.89|
|[phi-2-orange](https://huggingface.co/rhysjones/phi-2-orange)|  33.4|  71.3|     49.9|    37.3|  47.97|
|[phi-2](https://huggingface.co/microsoft/phi-2)|  28.0|   70.8|     44.4|   35.2|  44.61|

## 🧩 Configuration

```yaml
slices:
  - sources:
      - model: rhysjones/phi-2-orange
        layer_range: [0, 32]
      - model: cognitivecomputations/dolphin-2_6-phi-2
        layer_range: [0, 32]
merge_method: slerp
base_model: rhysjones/phi-2-orange
parameters:
  t:
    - filter: self_attn
      value: [0, 0.5, 0.3, 0.7, 1]
    - filter: mlp
      value: [1, 0.5, 0.7, 0.3, 0]
    - value: 0.5
dtype: bfloat16
```

## 💻 Usage

```python
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
torch.set_default_device("cuda")
model = AutoModelForCausalLM.from_pretrained("vince62s/phi-2-psy", torch_dtype="auto", trust_remote_code=True)
tokenizer = AutoTokenizer.from_pretrained("vince62s/phi-2-psy", trust_remote_code=True)
inputs = tokenizer('''def print_prime(n):
   """
   Print all primes between 1 and n
   """''', return_tensors="pt", return_attention_mask=False)
outputs = model.generate(**inputs, max_length=200)
text = tokenizer.batch_decode(outputs)[0]
print(text)
```