munish0838 commited on
Commit
bdf5b66
1 Parent(s): 22e6f60

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +65 -0
README.md ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: jdqwoi/TooManyMixRolePlay-7B-Story_V3.5
3
+ tags:
4
+ - merge
5
+ - mergekit
6
+ - lazymergekit
7
+ - jdqwoi/TooManyMixRolePlay-7B-Story_V2
8
+ - jdqwoi/TooManyMixRolePlay-7B-Story_V3
9
+ pipeline_tag: text-generation
10
+ ---
11
+
12
+ # TooManyMixRolePlay-7B-Story_V3.5-GGUF
13
+ This is quantized version of [jdqwoi/TooManyMixRolePlay-7B-Story_V3.5](https://huggingface.co/jdqwoi/TooManyMixRolePlay-7B-Story_V3.5) created using llama.cpp
14
+
15
+ # Model Description
16
+
17
+ TooManyMixRolePlay-7B-Story_V3.5 is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
18
+ * [jdqwoi/TooManyMixRolePlay-7B-Story_V2](https://huggingface.co/jdqwoi/TooManyMixRolePlay-7B-Story_V2)
19
+ * [jdqwoi/TooManyMixRolePlay-7B-Story_V3](https://huggingface.co/jdqwoi/TooManyMixRolePlay-7B-Story_V3)
20
+
21
+ ## 🧩 Configuration
22
+
23
+ ```yaml
24
+ slices:
25
+ - sources:
26
+ - model: jdqwoi/TooManyMixRolePlay-7B-Story_V2
27
+ layer_range: [0, 32]
28
+ - model: jdqwoi/TooManyMixRolePlay-7B-Story_V3
29
+ layer_range: [0, 32]
30
+ merge_method: slerp
31
+ base_model: jdqwoi/TooManyMixRolePlay-7B-Story_V2
32
+ parameters:
33
+ t:
34
+ - filter: self_attn
35
+ value: [0, 0.5, 0.3, 0.7, 1]
36
+ - filter: mlp
37
+ value: [1, 0.5, 0.7, 0.3, 0]
38
+ - value: 0.5
39
+ dtype: bfloat16
40
+ ```
41
+
42
+ ## 💻 Usage
43
+
44
+ ```python
45
+ !pip install -qU transformers accelerate
46
+
47
+ from transformers import AutoTokenizer
48
+ import transformers
49
+ import torch
50
+
51
+ model = "jdqwoi/TooManyMixRolePlay-7B-Story_V3.5"
52
+ messages = [{"role": "user", "content": "What is a large language model?"}]
53
+
54
+ tokenizer = AutoTokenizer.from_pretrained(model)
55
+ prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
56
+ pipeline = transformers.pipeline(
57
+ "text-generation",
58
+ model=model,
59
+ torch_dtype=torch.float16,
60
+ device_map="auto",
61
+ )
62
+
63
+ outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
64
+ print(outputs[0]["generated_text"])
65
+ ```