retiredcarboxyl
commited on
Commit
•
d0cbaf6
1
Parent(s):
99107e3
updated readme.md file and made some changes
Browse files
README.md
CHANGED
@@ -1,20 +1,19 @@
|
|
1 |
---
|
2 |
tags:
|
3 |
- merge
|
4 |
-
-
|
5 |
-
-
|
6 |
-
-
|
7 |
-
-
|
8 |
base_model:
|
9 |
- meta-llama/Llama-2-7b-chat-hf
|
10 |
-
- meta-llama/Llama-2-7b-hf
|
11 |
language:
|
12 |
- en
|
13 |
---
|
14 |
|
15 |
-
#
|
16 |
|
17 |
-
|
18 |
* [meta-llama/Llama-2-70b-chat-hf](https://huggingface.co/meta-llama/Llama-2-70b-chat-hf)
|
19 |
* [mistralai/Mixtral-8x7B-Instruct-v0.1](https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1)
|
20 |
* [google/gemma-7b](https://huggingface.co/google/gemma-7b)
|
@@ -25,12 +24,17 @@ CarbonGPT is a merge of the following models using [CarbonGPTMerger](https://col
|
|
25 |
```yaml
|
26 |
slices:
|
27 |
- sources:
|
28 |
-
- model: meta-llama/Llama-2-
|
29 |
layer_range: [0, 32]
|
30 |
-
- model:
|
31 |
layer_range: [0, 32]
|
|
|
|
|
|
|
|
|
|
|
32 |
merge_method: slerp
|
33 |
-
base_model: meta-llama/Llama-2-
|
34 |
parameters:
|
35 |
t:
|
36 |
- filter: self_attn
|
@@ -50,7 +54,7 @@ from transformers import AutoTokenizer
|
|
50 |
import transformers
|
51 |
import torch
|
52 |
|
53 |
-
model = "retiredcarboxyl/
|
54 |
messages = [{"role": "user", "content": "What is a large language model?"}]
|
55 |
|
56 |
tokenizer = AutoTokenizer.from_pretrained(model)
|
|
|
1 |
---
|
2 |
tags:
|
3 |
- merge
|
4 |
+
- meta-llama/Llama-2-70b-chat-hf
|
5 |
+
- mistralai/Mixtral-8x7B-Instruct-v0.1
|
6 |
+
- google/gemma-7b
|
7 |
+
- tiiuae/falcon-180B
|
8 |
base_model:
|
9 |
- meta-llama/Llama-2-7b-chat-hf
|
|
|
10 |
language:
|
11 |
- en
|
12 |
---
|
13 |
|
14 |
+
# Chat2Eco
|
15 |
|
16 |
+
Chat2Eco is a merge of the following models using [Chat2Eco](https://colab.research.google.com/drive/1ECBKKGQeV3OhnpiIaThZEl8XMQaXNoEh?usp=sharing):
|
17 |
* [meta-llama/Llama-2-70b-chat-hf](https://huggingface.co/meta-llama/Llama-2-70b-chat-hf)
|
18 |
* [mistralai/Mixtral-8x7B-Instruct-v0.1](https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1)
|
19 |
* [google/gemma-7b](https://huggingface.co/google/gemma-7b)
|
|
|
24 |
```yaml
|
25 |
slices:
|
26 |
- sources:
|
27 |
+
- model: meta-llama/Llama-2-70b-chat-hf
|
28 |
layer_range: [0, 32]
|
29 |
+
- model: mistralai/Mixtral-8x7B-Instruct-v0.1
|
30 |
layer_range: [0, 32]
|
31 |
+
- model: google/gemma-7b
|
32 |
+
layer_range: [0, 64]
|
33 |
+
- model: tiiuae/falcon-180B
|
34 |
+
layer_range: [0, 64]
|
35 |
+
|
36 |
merge_method: slerp
|
37 |
+
base_model: meta-llama/Llama-2-70b-chat-hf
|
38 |
parameters:
|
39 |
t:
|
40 |
- filter: self_attn
|
|
|
54 |
import transformers
|
55 |
import torch
|
56 |
|
57 |
+
model = "retiredcarboxyl/Chat2Eco"
|
58 |
messages = [{"role": "user", "content": "What is a large language model?"}]
|
59 |
|
60 |
tokenizer = AutoTokenizer.from_pretrained(model)
|