Commit
•
47d5ceb
1
Parent(s):
3431cbb
Update README.md
Browse files
README.md
CHANGED
@@ -106,6 +106,37 @@ This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/g
|
|
106 |
It achieves the following results on the evaluation set:
|
107 |
- Loss: 1.1468
|
108 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
109 |
## Model description
|
110 |
|
111 |
More information needed
|
|
|
106 |
It achieves the following results on the evaluation set:
|
107 |
- Loss: 1.1468
|
108 |
|
109 |
+
## How to use
|
110 |
+
|
111 |
+
**PEFT**
|
112 |
+
```python
|
113 |
+
from peft import PeftModel, PeftConfig
|
114 |
+
from transformers import AutoModelForCausalLM
|
115 |
+
|
116 |
+
model_id = "MaziyarPanahi/gemma-7b-alpaca-52k-v0.1"
|
117 |
+
|
118 |
+
config = PeftConfig.from_pretrained(model_id)
|
119 |
+
model = AutoModelForCausalLM.from_pretrained("google/gemma-7b")
|
120 |
+
model = PeftModel.from_pretrained(model, model_id)
|
121 |
+
```
|
122 |
+
|
123 |
+
**Transformers**
|
124 |
+
```python
|
125 |
+
# Use a pipeline as a high-level helper
|
126 |
+
from transformers import pipeline
|
127 |
+
|
128 |
+
model_id = "MaziyarPanahi/gemma-7b-alpaca-52k-v0.1"
|
129 |
+
|
130 |
+
pipe = pipeline("text-generation", model=model_id)
|
131 |
+
|
132 |
+
# Load model directly
|
133 |
+
from transformers import AutoTokenizer, AutoModelForCausalLM
|
134 |
+
|
135 |
+
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
136 |
+
model = AutoModelForCausalLM.from_pretrained(model_id)
|
137 |
+
```
|
138 |
+
|
139 |
+
|
140 |
## Model description
|
141 |
|
142 |
More information needed
|